text
stringlengths 204
3.13k
|
|---|
imports of goods and services and payments for investments and transfers. to England might purchase insurance from a British insurance company. A Dutch flower grower may fly flowers to the United States aboard a U.S. airliner. In the first case, the United States is importing services and therefore using up foreign exchange; in the second case, it is selling services to foreigners and earning foreign exchange. In 2017 the United States exported $780.9 billion in services and imported $538.1 billion, thus earning more foreign exchange than it used up regarding trade in services. The difference between a country’s exports of goods and services and its imports of goods and services is its balance of trade. When exports of goods and services are less than imports of goods and services, a country has a trade deficit. Table 19.1 shows that the U.S. trade deficit in 2017 was fairly large at $568.4 billion. Next in Table 19.1 comes investment income. U.S. citizens hold foreign assets (stocks, bonds, and real assets such as buildings and factories). Dividends, interest, rent, and profits paid to U.S. asset holders are a source of foreign exchange. Conversely, when foreigners earn dividends, interest, and profits on assets held in the United States, foreign exchange is used up. In 2017 the United States earned $926.9 in investment income and paid out $709.9 billion. Last in the current account are transfer payments. Transfer payments from the United States to foreigners are another use of foreign exchange. Some of these transfer payments are from private U.S. citizens, and some are from the U.S. government. You may send a check to a relief agency in Africa. Many immigrants in the United States send remittances to their countries of origin to help support extended families. Conversely, foreigners make transfer payments to the United States, which earns income for the United States. In 2017 the United States received $149.7 billion in transfer payments from abroad and sent $264.5 billion abroad. Line (10) in Table 19.1 shows the balance on current account. This is the balance of trade plus investment and transfer income and minus investment and transfer payments. Put another way, the balance on current account is the sum of income from exports of goods and services and income from investments and transfers minus payments for imports of goods and services and payments for investments and transfers. The balance on current account shows how much a nation has spent on foreign goods, services,
|
investment income payments, and transfers relative to how much it has earned from other countries. When the balance is negative, which it was for the United States in 2017, a nation has spent more on foreign goods and services (plus investment income and transfers paid) than it has earned through the sales of its goods and services to the rest of the world (plus investment income and transfers received). TABLE 19.1 U.S. Balance of Payments, 2017 Current Account Billions of dollars (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Goods exports Goods imports Exports of services Imports of services Balance of trade: (1) - (2) + (3) - (4) Investment income Investment payments Transfer income Transfer payments Balance on current account: (5) + (6) - (7) + (8) - (9) Financial account (11) Net capital transfer receipts Change in net U.S. liabilities (12) (13) Net receipts from financial derivatives (14) (15) Statistical discrepancy Balance of payments: (10) + (11) + (12) + (13) + (14) $1,550.7 2361.9 780.9 538.1 -568.4 926.9 709.0 149.7 264.5 -466.2 24.8 375.5 -26.4 92.2 0.0 Item (13) is the change in foreign assets in the United States minus the change in U.S. assets abroad. In 2017 this number was positive, which means that there was an increase in net U.S. liabilities. Source: Bureau of Economic Analysis, March 21, 2018. M19_CASE3826_13_GE_C19.indd 387 17/04/19 4:26 AM 388 PART V The World Economy The Financial Account MyLab Economics Concept Check For each transaction recorded in the current account, there is an offsetting transaction recorded in the financial account. Consider, for example, the $466.2 billion current account deficit that the United States ran in 2017. This deficit must be paid for, and how it is paid shows up in the financial account. The first two lines under the financial account in Table 19.1 are receipts recorded in the financial account: net capital transfer receipts and net receipts from financial derivatives. These are small. The first is positive, which says that the net
|
flow to the United States was positive. The second is negative, which was a net flow out of the United States. The third line, line (13), shows that net U.S. liabilities (to the rest of the world) increased by $375.5 billion. So the United States borrowed from the rest of the world (on net) $375.5 billion to partly finance the $466.2 billion deficit. If there were no measurement errors, the entire deficit would be financed by lines (11), (12), and (13): net capital receipts and net borrowing. There are, however, measurement errors, where the total error is called the statistical discrepancy. In 2017 the statistical discrepancy was $92.2 billion. This is, of course, a large error. But the main point to take away from this analysis is that aside from measurement errors, a current account deficit must be financed by changes in a country’s net capital receipts and its net liabilities to the rest of the world. The balance of payments—line (15) in Table 19.1—is always zero. An example may help in seeing the link between the current and financial accounts. Say a U.S. citizen buys a beer in a store on Caye Caulker, Belize, for $1.75 using U.S. currency, which is accepted in Belize along with the local currency. This is an import of the United States, so the U.S. currentaccount deficit has increased by $1.75. What happens on the financial account? The Belize store owner now has the $1.75, which is an asset for her (i.e., for Belize) and a liability for the United States. Net U.S. liabilities to the rest of the world have thus increased by $1.75—line (13) in Table 19.1. There are many international financial transactions that do not lead to a change in net U.S. liabilities in the financial account. If the Chinese central bank buys a U.S. government bond with yuan, its U.S. assets have increased (the bond), but so has its foreign liabilities (the yuan). In the United States there is an increase in foreign liabilities (the bond), but also an increase in foreign assets (the yuan). The net position of each country has not changed. The only way the net position of a country can change is through a positive or negative value of its current account. If in the
|
Belize example the U.S. citizen had simply exchanged $1.75 U.S. for $3.50 Belize (the exchange rate between the Belize dollar and the U.S. dollar is two to one), with no beer purchased, this would not have led to a change in net U.S. liabilities. This is just a swap of assets with no change in the current account. The balance of payments pertains to flows. In Table 19.1 these are flows for the year 2017. Regarding stocks, we know that stocks and flows are related. For example, the net wealth of a country vis-à-vis the rest of the world at the end of a given year is equal to its net wealth at the end of the previous year plus its current-account balance during the year. In 2017 the net wealth of the United States vis-à-vis the rest of the world decreased by $466.2 billion—its current account deficit for 2017. It is important to realize that the only way a country’s net wealth position can change is if its current account balance is nonzero. Simply switching one form of asset for another does not change a country’s net wealth position. A country’s net wealth position is simply the sum of all its past current account balances. Prior to the mid-1970s, the United States had generally run current account surpluses, and thus its net wealth position was positive. It was a creditor nation. This began to turn around in the mid-1970s, and by the mid-1980s, the United States was running large current account deficits. Sometime during this period, the United States changed from having a positive net wealth position vis-à-vis the rest of the world to having a negative position. In other words, the United States changed from a creditor nation to a debtor nation. The current account deficits have persisted, and the United States is now the largest debtor nation in the world. At the end of 2017 foreign assets in the United States totaled $35.5 trillion and U.S. assets abroad totaled $27.6 trillion.2 The U.S. net wealth position was thus -$7.9 trillion. This large negative position reflects the fact that the United States has spent much more since the 1970s on foreign goods and services (plus investment income and transfers paid) than it earned through the sales of its goods and services to the rest of the world (plus investment income and transfers received
|
). 2Bureau of Economic Analysis, March 31, 2015. M19_CASE3826_13_GE_C19.indd 388 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 389 Debtor and Creditor Nations Since their independence, developing nations have been net recipients of foreign debt from developed nations to finance their development plans. Many of the poorest countries of the world are overwhelmed with debt burdens that they find difficult to manage. This has urged the International Monetary Fund (IMF) and the World Bank to start the Heavily Indebted Poor Countries (HIPC) Initiative in 1996. To help with economic and human development, the HIPC Initiative has extended $99 billion in the form of debt relief and lowinterest loans to 37 very poor nations. However, since the beginning of the twenty-first century, several industrial nations have become debtor nations instead of their previous role as suppliers of capital to developing nations. In response to a slowdown in the growth prospects of industrial nations, middle-income and emerging market economies have been acting as their net creditors, especially during the global financial crisis of 2008–2009. The United Nations reports that external debt levels of high-income countries are on average 2 to 3 times higher than in low and middle-income developing nations. As such, the net financial flows to developing countries turned negative between 2015 and 2017. Since 2006, with this change in global capital flows, the IMF started building a database of the balance of payments figures of countries round the world. Among these are the Net International Investment Position (NIIP), which is defined as the difference between foreign assets that domestic residents own and liabilities. When external assets exceed liabilities, the NIIP is positive and when liabilities exceed assets it is negative. The United States is the largest debtor nation in the world, with a negative NIIP of €7,710.5 billion. For the same period, Japan recorded a positive NIIP of €2,829.1 billion. The European Union (EU-28) reported a negative NIIP of nearly €500 billion in 2017, which equals almost 5 percent of GDP. Germany and the Netherlands are the major net creditor economies in Europe, while Spain, Ireland, and France held higher positions in financial liabilities abroad than financial assets, making them net borrowers or debtors.1 Thus, the risks of debt sustainability have grown for a few developed countries as well as for some
|
developing and emerging economies. The NIIP position is an important parameter to gauge a country’s financial condition and creditworthiness. But instead of looking at absolute values, the NIIP’s size has to be examined in relation to the economy’s size (the ratio of NIIP to GDP). CRITICAL THINKING 1. Visit the IMF database at www.data.imf.org. Examine the NIIP position of your country in relation to its GDP. Is your country a net debtor or creditor? 1European Central Bank (2018), “Euro Area Quarterly Balance of Payments and International Investment Position,” January 11. Equilibrium Output (Income) in an Open Economy Everything we have said so far has been descriptive. Now we turn to analysis. How are all these trade and capital flows determined? What impacts do they have on the economies of the countries involved? To simplify our discussion, we will assume that exchange rates are fixed. We will relax this assumption later. 19.2 LEARNING OBJECTIVE Discuss how equilibrium output is determined in an open economy, and describe the trade feedback effect and the price feedback effect. The International Sector and Planned Aggregate Expenditure MyLab Economics Concept Check The first change we will have to make to take into account the openness of the economy is in the calculation of the multiplier, one of the backbones of economic policy analysis. Our earlier calculations of the multiplier defined aggregate expenditure (AE) as consisting of the consumption of households (C), the planned investment of firms (I), and the spending of the government (G). With an open economy, we must now include in aggregate expenditures the goods and services a country exports to the rest of the world, EX, and we will also have to make an adjustment M19_CASE3826_13_GE_C19.indd 389 17/04/19 4:26 AM 390 PART V The World Economy for what it imports, IM. Clearly EX should be included as part of total output and income. A U.S. razor sold to a buyer in Mexico is as much a part of U.S. production as a similar razor sold in Pittsburgh. Exports simply represent demand for domestic products not by domestic households and firms and the government, but by the rest of the world. What about imports? Imports are not a part of domestic output (Y) because they are produced outside the home country. When we calculate households’ total consumption
|
spending, firms’ total investment spending, and total government spending, imports are included. Therefore, to calculate domestic output correctly, we must subtract the parts of consumption, investment, and government spending that constitute imports. The definition of planned aggregate expenditure becomes: Planned aggregate expenditure in an open economy: AE K C + I + G + EX - IM net exports of goods and services (EX - IM) The difference between a country’s total exports and total imports. The last two terms (EX - IM) together are the country’s net exports of goods and services. Determining the Level of Imports What determines the level of imports and exports in a country? Clearly the level of imports is a function of income (Y). When U.S. income increases, U.S. citizens buy more of everything, including Japanese cars and Korean smartphones. When income rises, imports tend to go up. Algebraically, IM = mY marginal propensity to import (MPM) The change in imports caused by a $1 change in income. where Y is income and m is some positive number. (m is assumed to be less than one; otherwise, a $1 increase in income generates an increase in imports of more than $1, which is unrealistic.) Recall from Chapter 8 that the marginal propensity to consume (MPC) measures the change in consumption that results from a $1 change in income. Similarly, the marginal propensity to import, abbreviated as MPM or m, is the change in imports caused by a $1 change in income. If m = 0.2, or 20 percent, and income is $1,000, then imports, IM, are equal to 0.2 * +1,000 = +200. If income rises by $100 to $1,100, the change in imports will equal m * = 0.2 * +100 = +20. the change in income For now we will assume that exports (EX) are given (that is, they are not affected, even indi- rectly, by the state of the domestic economy.) This assumption is relaxed later in this chapter. 1 2 Solving for Equilibrium Given the assumption about how imports are determined, we can solve for equilibrium income. This procedure is illustrated in Figure 19.1. Starting from the consumption function (blue line) in Figure 19.1(a), we gradually build up the components of planned aggregate expenditure (red line). Assuming for simplicity that planned investment, government purchases,
|
and exports are all constant and do not depend on income, we move easily from the blue line to the red line by adding the fixed amounts of I, G, and EX to consumption at every level of income. In this example, we take I + G + EX to equal 80. C + I + G + EX, however, includes spending on imports, which are not part of domestic production. To get spending on domestically produced goods, we must subtract the amount that is imported at each level of income. In Figure 19.1(b), we assume m = 0.25, so that 25 percent of total income is spent on goods and services produced in foreign countries. For example, at Y = 200, IM = 0.25 Y, or 50. Similarly, at Y = 400, IM = 0.25 Y, or 100. Figure 19.1(b) shows the planned domestic aggregate expenditure curve that nets out imports from expenditures. Equilibrium is reached when planned domestic aggregate expenditure equals domestic aggregate output (income). This is true at only one level of aggregate output, Y* = 200, in Figure 19.1(b). If Y were below Y*, planned expenditure would exceed output, inventories would be lower than planned, and output would rise. At levels above Y*, output would exceed planned expenditure, inventories would be larger than planned, and output would fall. The Open-Economy Multiplier All of this has implications for the size of the multiplier. Recall the multiplier, introduced in Chapter 8, and consider a sustained rise in government purchases (G). Initially, the increase in G will cause planned aggregate expenditure to be greater than aggregate output. Domestic firms will find their inventories to be lower than planned and thus will increase M19_CASE3826_13_GE_C19.indd 390 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 391 ) $ ( 400 350 300 250 200 150 100 50 Planned aggregate expenditure ; C 1 I 1 G 1 EX I + G + EX = 80 a. 458 Consumption function ) $ ( 350 300 250 200 150 100 50 b. 458 Planned aggregate expenditure ; C 1 I 1 G 1 EX IM 5.25Y 5 100 IM 5.25Y 5 50 Planned domestic aggregate expenditure ; C 1 I 1 G 1 EX 2 IM 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Aggregate output (income) (Y
|
) MyLab Economics Concept Check Aggregate output (income) (Y) Y* ▴ FIGURE 19.1 Determining Equilibrium Output in an Open Economy In a., planned investment spending (I), government spending (G), and total exports (EX) are added to consumption (C) to arrive at planned aggregate expenditure. However, C + I + G + EX includes spending on imports. In b., the amount imported at every level of income is subtracted from planned aggregate expenditure. Equilibrium output occurs at Y* = 200, the point at which planned domestic aggregate expenditure crosses the 45-degree line. their output, but added output means more income. More workers are hired, and profits are higher. Some of the added income is saved, and some is spent. The added consumption spending leads to a second round of inventories being lower than planned and raising output. Equilibrium output rises by a multiple of the initial increase in government purchases. This is the multiplier effect. In Chapters 8 and 9, we showed that the simple multiplier equals 1 MPS). That is, a sustained increase in government purchases equal to ΔG will lead to an increase in >1 aggregate output (income) of ΔG. If the MPC were 0.75 and government purchases rose by $10 billion, equilibrium income would rise by 4 * $10 billion, or $40 billion. The 24 multiplier is 1 - MPC 1 - MPC 3 >1 025, or (1 1 1 1 2 > 1 - 0.75 = 4.0. = > 3 3 4 24 >1 In an open economy, some of the increase in income brought about by the increase in G is spent on imports instead of domestically produced goods and services. The part of income spent on imports does not increase domestic income (Y) because imports are produced by foreigners. To compute the multiplier, we need to know how much of the increased income is used to increase domestic consumption. (We are assuming all imports are consumption goods. In practice, some imports are investment goods and some are goods purchased by the government.) In other words, we need to know the marginal propensity to consume domestically produced goods. Domestic consumption is C - IM. So the marginal propensity to consume domestic goods is the marginal propensity to consume all goods (the MPC) minus the marginal propensity to import (the MPM). The marginal propensity to consume domestic goods is. Consequently, MPC - MPM 1 open@economy multipler = 1 2 1 - MPC
|
- MPM 2 1 If the MPC is 0.75 and the MPM is 0.25, then the multiplier is 1/0.5, or 2.0. This multiplier is smaller than the multiplier in which imports are not taken into account, which is 1/0.25, or 4.0. The effect of a sustained increase in government spending (or investment) on income— that is, the multiplier—is smaller in an open economy than in a closed economy. The reason: When government spending (or investment) increases and income and consumption rise, some of the extra consumption spending that results is on foreign products and not on domestically produced goods and services. In an open economy the impact of government spending on the domestic economy is less than it otherwise would be. At the same time, one country’s government spending increases affects other countries. Fiscal policy in one country can affect the macroeconomy is its trading partners. We see from this that taking imports into account reduces the multiplier. This effect is especially large in the case of small, open economies like the Caribbean nations in which much of what is consumed is imported. As we will see, in these countries, the government has very little independent control over fiscal policy. M19_CASE3826_13_GE_C19.indd 391 17/04/19 4:26 AM 392 PART V The World Economy Imports, Exports, and the Trade Feedback Effect MyLab Economics Concept Check For simplicity, we have so far assumed that the level of imports depends only on income and that the level of exports is fixed. In reality, the amount of spending on imports also depends on factors other than income and exports are not fixed. We will now consider the more realistic picture. The Determinants of Imports The same factors that affect households’ consumption behavior and firms’ investment behavior are likely to affect the demand for imports because some imported goods are consumption goods and some are investment goods. For example, anything that increases consumption spending is likely to increase the demand for imports. We saw in Chapters 8 and 9 that factors such as the after-tax real wage, after-tax nonlabor income, and interest rates affect consumption spending; thus, they should also affect spending on imports. Similarly, anything that increases investment spending is likely to increase the demand for imports. A decrease in interest rates, for example, should encourage spending on both domestically produced goods and foreign-produced goods. There is one additional consideration in determining spending on
|
imports: the relative prices of domestically produced and foreign-produced goods. If the prices of foreign goods fall relative to the prices of domestic goods, people will consume more foreign goods relative to domestic goods. When Japanese cars are inexpensive relative to U.S. cars, consumption of Japanese cars should be high and vice versa. The Determinants of Exports We now relax our assumption that exports are fixed. The foreign demand for U.S. exports is identical to the foreign countries’ imports from the United States. Germany imports goods, some of which are U.S.-produced. Total expenditure on imports in Germany is a function of the factors we just discussed except that the variables are German variables instead of U.S. variables. This is true for all other countries as well. The demand for U.S. exports thus depends on economic activity in the rest of the world—rest-of-the-world real wages, wealth, nonlabor income, interest rates, and so forth—as well as on the prices of U.S. goods relative to the prices of rest-of-the-world goods. When foreign output increases, U.S. exports tend to increase. In this way economic growth in the rest of the world stimulates the economy in the United States. U.S. exports also tend to increase when U.S. prices fall relative to foreign goods prices. With an open economy, countries are interdependent. U.S. exports also tend to increase when U.S. prices fall relative to foreign prices. The Trade Feedback Effect We can now combine what we know about the demand for imports and the demand for exports to discuss the trade feedback effect. Suppose the United States finds its exports increasing, perhaps because the world suddenly decides it prefers U.S. computers to other computers. Rising exports will lead to an increase in U.S. output (income), which leads to an increase in U.S. imports. Here is where the trade feedback begins. U.S. imports are somebody else’s exports. The extra import demand from the United States raises the exports of the rest of the world. When other countries’ exports to the United States go up, their output and incomes also rise, in turn leading to an increase in the demand for imports from the rest of the world. Some of the extra imports demanded by the rest of the world come from the United States, so U.S. exports increase. The increase in U.S. exports stimulates U.S
|
. economic activity even more, triggering a further increase in the U.S. demand for imports and so on. An increase in U.S. imports increases other countries’ exports, which stimulates those countries’ economies and increases their imports, which increases U.S. exports, and so on. This is the trade feedback effect. In other words, an increase in U.S. economic activity leads to a worldwide increase in economic activity, which then “feeds back” to the United States. Import and Export Prices and the Price Feedback Effect MyLab Economics Concept Check We have talked about the price of imports, but we have not yet discussed the factors that influence import prices. The consideration of import prices is complicated because more than one currency is involved. When we talk about “the price of imports,” do we mean the trade feedback effect The tendency for an increase in the economic activity of one country to lead to a worldwide increase in economic activity, which then feeds back to that country. M19_CASE3826_13_GE_C19.indd 392 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 393 price in dollars, in yen, or in euros? The same question holds for the price of exports because the exports of one country are the imports of another. When Mexico exports auto parts to the United States, Mexican manufacturers are interested in the price of auto parts in terms of pesos because pesos are what they use for transactions in Mexico. U.S. consumers are interested in the price of auto parts in dollars because dollars are what they use for transactions in the United States. The link between the two prices is the dollar/peso exchange rate. Suppose Mexico is experiencing inflation and the price of radiators in pesos rises from 1,000 pesos to 1,200 pesos per radiator. If the dollar/peso exchange rate remains unchanged at, say, $0.10 per peso, Mexico’s export price for radiators in terms of dollars will also rise from $100 to $120 per radiator. Because Mexico’s exports to the United States are, by definition, U.S. imports from Mexico, an increase in the dollar prices of Mexican exports to the United States means an increase in the prices of U.S. imports from Mexico. Therefore, when Mexico’s export prices rise with no change in the dollar/pes
|
o exchange rate, U.S. import prices rise. Export prices of other countries affect U.S. import prices. A country’s export prices tend to move fairly closely with the general price level in that country. If Mexico is experiencing a general increase in prices, this change likely will be reflected in price increases of all domestically produced goods, both exportable and nonexportable. The general rate of inflation abroad is likely to affect U.S. import prices. If the inflation rate abroad is high, U.S. import prices are likely to rise. The Price Feedback Effect We have just seen that when a country experiences an increase in domestic prices, the prices of its exports will increase. It is also true that when the prices of a country’s imports increase, the prices of domestic goods may increase in response. There are at least two ways this effect can occur. First, an increase in the prices of imported inputs will increase the costs of firms which use these imports as inputs, causing a country’s aggregate supply curve to shift to the left. Recall from Chapter 12 that a leftward shift in the aggregate supply curve resulting from a cost increase causes aggregate output to fall and prices to rise (stagflation). Second, if import prices rise relative to domestic prices, households will tend to substitute domestically produced goods and services for imports. This is equivalent to a rightward shift of the aggregate demand curve. If the domestic economy is operating on the upward-sloping part of the aggregate supply curve, the overall domestic price level will rise in response to an increase in aggregate demand. Perfectly competitive firms will see market-determined prices rise, and imperfectly competitive firms will experience an increase in the demand for their products. Studies have shown, for example, that the price of automobiles produced in the United States moves closely with the price of imported cars. Still, this is not the end of the story. Suppose a country—say, Mexico—experiences an increase in its domestic price level. This will increase the price of its exports to Canada (and to all other countries). The increase in the price of Canadian imports from Mexico will lead to an increase in domestic prices in Canada. Canada also exports to Mexico. The increase in Canadian prices causes an increase in the price of Canadian exports to Mexico, which then further increases the Mexican price level. This is called the price feedback effect, in the sense that inflation is “exportable.” An increase in the price level in one country can drive up
|
prices in other countries, which in turn further increases the price level in the first country. Through export and import prices, a domestic price increase can “feed back” on itself. It is important to realize that the discussion so far has been based on the assumption of fixed exchange rates. Life is more complicated under flexible exchange rates, to which we now turn. price feedback effect The process by which a domestic price increase in one country can “feed back” on itself through export and import prices. An increase in the price level in one country can drive up prices in other countries. This in turn further increases the price level in the first country. The Open Economy with Flexible Exchange Rates Exchange rates are a factor in determining the flow of international trade and the structure of those exchange rates thus matters. In practice, the structure of exchange rates has changed considerably over time, influenced in part by international agreements and events. 19.3 LEARNING OBJECTIVE Discuss factors that affect exchange rates in an open economy with a floating system. M19_CASE3826_13_GE_C19.indd 393 17/04/19 4:26 AM 394 PART V The World Economy floating, or marketdetermined, exchange rates Exchange rates that are determined by the unregulated forces of supply and demand. In the early part of the twentieth century, nearly all currencies were backed by gold. Their values were fixed in terms of a specific number of ounces of gold, which determined their values in international trading—exchange rates. In 1944, with the international monetary system in chaos as the end of World War II drew near, a large group of experts unofficially representing 44 countries met in Bretton Woods, New Hampshire, and drew up a number of agreements. One of those agreements established a system of essentially fixed exchange rates under which each country agreed to intervene by buying and selling currencies in the foreign exchange market when necessary to maintain the agreed-to value of its currency. In 1971, most countries, including the United States, began to allow exchange rates to be flexible, determined essentially by supply and demand. Flexible exchange rates are known as floating or market determined exchange rates. Although there are considerable intricacies in the way flexible exchange rates operate, the logic is straightforward. If British goods are popular with U.S. consumers, there will be a large demand for British pounds by the U.S. customers for those goods. If the British don’t like U.S. goods, few will
|
demand U.S. dollars to buy those goods. Without government intervention in the marketplace, the price of British pounds in dollars would rise in this situation as those who want to exchange dollars for pounds (those who “demand” pounds) exceed those who want to exchange pounds for dollars (those who “supply” pounds). The exchange rate market thus reflects the markets for real goods in the various economies. Although governments still intervene to ensure that exchange rate movements are “orderly,” exchange rates today are largely determined by the unregulated forces of supply and demand. Understanding how an economy interacts with the rest of the world when exchange rates are not fixed is not as simple as when we assume fixed exchange rates. Exchange rates determine the price of imported goods relative to domestic goods and can have significant effects on the level of imports and exports. Consider a 20 percent drop in the value of the dollar against the British pound. Dollars buy fewer pounds, and pounds buy more dollars. Both British residents, who now get more dollars for pounds, and U.S. residents, who get fewer pounds for dollars, find that U.S. goods and services are more attractive. Exchange rate movements have important impacts on imports, exports, and the movement of capital between countries. The Market for Foreign Exchange MyLab Economics Concept Check What determines exchange rates under a floating rate system? To explore this question, we assume that there are just two countries, the United States and Great Britain. It is easier to understand a world with only two countries, and most of the points we will make can be generalized to a world with many trading partners. The Supply of and Demand for Pounds Governments, private citizens, banks, and corporations exchange pounds for dollars and dollars for pounds every day. In our two-country case, those who demand pounds are holders of dollars seeking to exchange them for pounds to buy British goods, travel to Britain, or invest in British stocks and bonds. Those who supply pounds are holders of pounds seeking to exchange them for dollars to buy U.S. goods, visit or invest in the United States. The supply of dollars on the foreign exchange market is the number of dollars that holders seek to exchange for pounds in a given time period. The demand for and supply of dollars on foreign exchange markets determine exchange rates. In addition to buyers and sellers who exchange money to engage in transactions, some people and institutions hold currency balances for speculative reasons. If you think that the U.S. dollar is going to decline in value relative to the
|
pound, you may want to hold some of your wealth in the form of pounds. Table 19.2 summarizes some of the major categories of private foreign exchange demanders and suppliers in the two-country case of the United States and Great Britain. We can use a variant of supply and demand analysis to help us understand the exchange rate in currency markets. Figure 19.2 shows the demand curve for pounds in the foreign exchange market. On the vertical axis is the price of pounds, expressed in dollars per pound, and on the horizontal axis is the quantity of pounds. Thus, as we move down the vertical axis, the pound depreciates relative to the dollar—it takes fewer dollars to buy a pound. Suppose we start at a point at which it costs $2 to buy one pound and the price of a British good is one pound. To buy that good at the existing exchange rate would cost an American $2. Let us suppose at that exchange rate 100 units of the good are demanded, giving rise to a demand for 100 pounds in the currency market. Now let the pound depreciate, so that it costs only $1 to buy a pound. It seems likely that the British good M19_CASE3826_13_GE_C19.indd 394 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 395 TABLE 19.2 Some Buyers and Sellers in International Exchange Markets: United States and Great Britain The Demand for Pounds (Supply of Dollars) 1. Firms, households, or governments that import British goods into the United States or want to buy British-made goods and services 2. U.S. citizens traveling in Great Britain 3. Holders of dollars who want to buy British stocks, bonds, or other financial instruments 4. U.S. companies that want to invest in Great Britain 5. Speculators who anticipate a decline in the value of the dollar relative to the pound The Supply of Pounds (Demand for Dollars) 1. Firms, households, or governments that import U.S. goods into Great Britain or want to buy U.S.-made goods and services 2. British citizens traveling in the United States 3. Holders of pounds who want to buy stocks, bonds, or other financial instruments in the United States 4. British companies that want to invest in the United States 5. Speculators who anticipate a rise in the value of the dollar relative to the pound
|
will look more attractive to Americans. With a fixed price in pounds for the good in question, Americans can buy that one pound good for only $1 rather than the original $2. Whereas people originally wanted 100 units of the good, with a dollar price much reduced they will likely want more than 100 units. To facilitate that transaction will thus require more than 100 pounds. The demand-for-pounds curve in the foreign exchange market thus has a negative slope. What about the supply of pounds? Pounds are supplied by the British who want to buy U.S. goods. Figure 19.3 shows a supply curve for pounds in the foreign exchange market. As we move up the vertical axis, the dollar becomes cheaper; each pound translates into more dollars, making the price of U.S.-produced goods and services lower to the British. The British buy more U.S.-made goods when the price of pounds is high (the value of the dollar is low). If the demand for U.S. imports is elastic then that increase in British demand for U.S. goods and services increases the quantity of pounds supplied. The curve representing the supply of pounds in the foreign exchange market has a positive slope. The key to understanding the supply and demand curves represented here is to recognize that the price on the vertical axis is the price of one currency relative to a second. As we go down the vertical axis in this case, the pound becomes less expensive relative to the dollar, or equivalently, the dollar becomes more expensive relative to the pound. Moving down the vertical axis to the origin, the low relative price of the pound induces demand for pounds by Americans ◂ FIGURE 19.2 The Demand for Pounds in the Foreign Exchange Market When the price of pounds falls, British-made goods and services appear less expensive to U.S. buyers. If British prices are constant, U.S. buyers will buy more British goods and services and the quantity of pounds demanded will rise Quantity of pounds, £ MyLab Economics Concept Check M19_CASE3826_13_GE_C19.indd 395 17/04/19 4:26 AM 396 PART V The World Economy ▸ FIGURE 19.3 The Supply of Pounds in the Foreign Exchange Market When the price of pounds rises, the British can obtain more dollars for each pound. This means that U.S.-made goods and services appear less expensive to British buyers. Thus, the quantity of pounds supplied is likely to rise with the exchange rate appreciation
|
of a currency The rise in the price of one currency relative to another. depreciation of a currency The fall in the price of one currency relative to another. ▸ FIGURE 19.4 The Equilibrium Exchange Rate When exchange rates are allowed to float, they are determined by the forces of supply and demand. An excess demand for pounds will cause the pound to appreciate against the dollar. An excess supply of pounds will lead to a depreciating pound. MyLab Economics Concept Check Quantity of pounds, £ to buy British goods. At the same time, however, British buyers are less interested in the now expensive American goods, and fewer pounds are supplied. The Equilibrium Exchange Rate When exchange rates are allowed to float, they are determined the same way other prices are determined: The equilibrium exchange rate occurs at the point at which the quantity demanded of a foreign currency equals the quantity of that currency supplied. This is illustrated in Figure 19.4. An excess demand for pounds (quantity demanded in excess of quantity supplied) will cause the price of pounds to rise—the pound will appreciate relative to the dollar. An excess supply of pounds will cause the price of pounds to fall—the pound will depreciate relative to the dollar.3 $1.50 ) £ / $ ( £* MyLab Economics Concept Check Quantity of pounds, £ 3Although Figure 19.3 shows the supply-of-pounds curve in the foreign exchange market with a positive slope, under certain circumstances the curve may bend back. Suppose the price of a pound rises from $1.50 to $2.00. Consider a British importer who buys 10 Chevrolets each month at $15,000 each, including transportation costs. When a pound exchanges for $1.50, he will supply 100,000 pounds per month to the foreign exchange market—100,000 pounds brings $150,000, enough to buy 10 cars. Now suppose the cheaper dollar causes him to buy 12 cars. Twelve cars will cost a total of $180,000; but at $2 = 1 pound, he will spend only 90,000 pounds per month. The supply of pounds on the market falls when the price of pounds rises. The reason for this seeming paradox is simple. The number of pounds a British importer needs to buy U.S. goods depends on both the quantity of goods he buys and the price of those goods in pounds. If demand for imports is inelastic so that the percentage decrease in price resulting from the depreciated currency is
|
greater than the percentage increase in the quantity of imports demanded, importers will spend fewer pounds and the quantity of pounds supplied in the foreign exchange market will fall. The supply of pounds will slope upward as long as the demand for U.S. imports is elastic. M19_CASE3826_13_GE_C19.indd 396 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 397 Factors That Affect Exchange Rates MyLab Economics Concept Check We now know enough to discuss the factors likely to influence exchange rates. Anything that changes the behavior of the people in Table 19.2 can cause demand and supply curves to shift and the exchange rate to adjust accordingly. Purchasing Power Parity: The Law of One Price If the costs of transporting goods between two countries are small, we would expect the price of the same good in both countries to be roughly the same. The price of basketballs should be roughly the same in Canada and the United States, for example. It is not hard to see why. If the price of basketballs is cheaper in Canada, it will benefit someone to buy balls in Canada at a low price and sell them in the United States at a higher price. This decreases the supply of basketballs in Canada and pushes up the price and increases the supply of balls in the United States, and pushes down the price. This process should continue as long as the price differential, and therefore the profit opportunity, persists. For a good with trivial transportation costs, we would expect this law of one price to hold. The price of a good should be the same regardless of where we buy it. Price differences across the two countries create arbitrage opportunities. If the law of one price held for all goods and if each country consumed the same market basket of goods, the exchange rate between the two currencies would be determined simply by the relative price levels in the two countries. If the price of a basketball were $10 in the United States and $12 in Canada, the U.S.–Canada exchange rate would have to be $1 U.S. per $1.20 Canadian. If the exchange rate were instead one-to-one, it would be profitable for people to buy the balls in the United States and sell them in Canada. This would increase the demand for U.S. dollars in Canada, thereby driving up their price in terms of Canadian dollars to $1 U.S. per $1.2 Canadian,
|
at which point no one could make a profit shipping basketballs across international lines and the process would cease.4 The theory that exchange rates will adjust so that the price of similar goods in different countries is the same is known as the purchasing-power-parity theory. According to this theory, if it takes 10 times as many Mexican pesos to buy a pound of salt in Mexico as it takes U.S. dollars to buy a pound of salt in the United States, the equilibrium exchange rate should be 10 pesos per dollar. In practice, transportation costs for many goods are quite large and the law of one price does not hold for these goods. (Haircuts are often cited as a good example. The transportation costs for a U.S. resident to get a British haircut are indeed large unless that person is an airline pilot.) Also, many products that are potential substitutes for each other are not precisely identical. For instance, a Rolls Royce and a Honda are both cars, but there is no reason to expect the exchange rate between the British pound and the yen to be set so that the prices of the two are equalized. In addition, countries consume different market baskets of goods, so we would not expect the aggregate price levels to follow the law of one price. Nevertheless, a high rate of inflation in one country relative to another puts pressure on the exchange rate between the two countries, and there is a general tendency for the currencies of relatively high-inflation countries to depreciate. Figure 19.5 shows the adjustment likely to occur following an increase in the U.S. price level relative to the price level in Great Britain. This change in relative prices will affect citizens of both countries. Higher prices in the United States make imports relatively less expensive. U.S. citizens are likely to increase their spending on imports from Britain, shifting the demand for pounds to the right, from D0 to D1. At the same time, the British see U.S. goods getting more expensive and reduce their demand for exports from the United States. Consequently, the supply of pounds shifts to the left, from S0 to S1. The result is an increase in the price of pounds. Before the change in relative prices, one pound sold for $1.50; after the change, one pound costs $2.00. The pound appreciates, and the dollar depreciates. Relative Interest Rates Another factor that influences a country’s exchange rate is the level of its interest rate relative to other countries
|
’ interest rates. If the interest rate is 2 percent in 4Of course, if the rate were $1 U.S. to $2 Canadian, it would benefit people to buy basketballs in Canada (at $12 Canadian, which is $6 U.S.) and sell them in the United States. This would weaken demand for the U.S. dollar, and its price would fall from $2 Canadian until it reached $1.20 Canadian. If the costs law of one price of transportation are small, the price of the same good in different countries should be roughly the same. purchasing-power-parity theory A theory of international exchange holding that exchange rates are set so that the price of similar goods in different countries is the same. M19_CASE3826_13_GE_C19.indd 397 17/04/19 4:26 AM 398 PART V The World Economy ▸ FIGURE 19.5 Exchange Rates Respond to Changes in Relative Prices The higher price level in the United States makes imports relatively less expensive. U.S. citizens are likely to increase their spending on imports from Britain, shifting the demand for pounds to the right, from D0 to D1. At the same time, the British see U.S. goods getting more expensive and reduce their demand for exports from the United States. The supply of pounds shifts to the left, from S0 to S1. The result is an increase in the price of pounds. The pound appreciates, and the dollar is worth less. $2.00 $1.50 ) £ / $ ( S1 S0 D1 D0 0 £* MyLab Economics Concept Check Quantity of pounds, £ the United States and 3 percent in Great Britain, people with money to lend have an incentive to buy British securities instead of U.S. securities. Although it is sometimes difficult for individuals in one country to buy securities in another country, it is easy for international banks and investment companies to do so. If the interest rate is lower in the United States than in Britain, there will be a movement of funds out of U.S. securities into British securities as banks and firms move their funds to the higher-yielding securities. How does a U.S. bank buy British securities? It takes its dollars, buys British pounds, and uses the pounds to buy the British securities. The bank’s purchase of pounds drives up the price of pounds in the foreign exchange market. The increased demand for pounds increases
|
the price of the pound (and decreases the price of the dollar). A high interest rate in Britain relative to the interest rate in the United States tends to depreciate the dollar. Figure 19.6 shows the effect of rising interest rates in the United States on the dollar-to-pound exchange rate. Higher interest rates in the United States attract British investors. To buy U.S. securities, the British need dollars. The supply of pounds (the demand for dollars) shifts to the right, from S0 to S1. The same relative interest rates affect the portfolio choices of U.S. banks, firms, and households. With higher interest rates at home, there is less incentive for U.S. residents to ▸ FIGURE 19.6 Exchange Rates Respond to Changes in Relative Interest Rates If U.S. interest rates rise relative to British interest rates, British citizens holding pounds may be attracted into the U.S. securities market. To buy bonds in the United States, British buyers must exchange pounds for dollars. The supply of pounds shifts to the right, from S0 to S1. At the same time, U.S. citizens are less likely to be interested in British securities because interest rates are higher at home. The demand for pounds shifts to the left, from D0 to D1. The result is the pound depreciates vis-a-vis the dollar and the dollar (naturally) appreciates visa-vis the pound. S0 S1 D0 D1 ) £ / $ ( 1.50 $1.00 0 £* MyLab Economics Concept Check Quantity of pounds, £ M19_CASE3826_13_GE_C19.indd 398 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 399 buy British securities. The demand for pounds drops at the same time the supply increases and the demand curve shifts to the left, from D0 to D1. The net result is a depreciating pound and an appreciating dollar. The price of pounds falls from $1.50 to $1.00. The Effects of Exchange Rates on the Economy MyLab Economics Concept Check We are now ready to discuss some of the implications of floating exchange rates. Recall, when exchange rates are fixed, households spend some of their incomes on imports and the multiplier is smaller than it would be otherwise. Imports are a “leakage” from the circular flow,
|
much like taxes and saving. Exports, in contrast, are an “injection” into the circular flow; they represent spending on U.S.-produced goods and services from abroad and can stimulate output. Exchange Rate Effects on Imports, Exports, and Real GDP As we already know, when a country’s currency depreciates (falls in value), its import prices rise and its export prices (in foreign currencies) fall. When the U.S. dollar is cheap, U.S. products are more competitive with products produced in the rest of the world and foreign-made goods look expensive to U.S. citizens. A depreciation of a country’s currency can thus serve as a stimulus to the economy. Suppose the U.S. dollar depreciates relative to the other major currencies, as it did sharply between 1985 and 1988 and again, more moderately from 2002 to 2008 and 2012 to 2013. If foreign buyers increase their spending on U.S. goods, and domestic buyers substitute U.S.-made goods for imports, aggregate expenditure on domestic output will rise, inventories will fall, and real GDP (Y) will increase. A depreciation of a country’s currency is likely to increase its GDP.5 Exchange Rates and the Balance of Trade: The J Curve A depreciating currency tends to increase exports and decrease imports, so you might think that it also will reduce a country’s trade deficit. In fact, the effect of depreciation on the balance of trade is ambiguous. Many economists believe that when a currency starts to depreciate, the balance of trade is likely to worsen for the first few quarters (perhaps three to six). After that, the balance of trade may improve. This effect is graphed in Figure 19.7. The curve in this figure resembles the letter J, and the movement in the balance of trade that it describes is sometimes called the J-curve effect. The point of the J shape is that the balance of trade gets worse before it gets better following a currency depreciation. How does the J curve come about? Recall from Table 19.1 that the balance of trade is equal to export revenue minus import costs, including exports and imports of services: balance of trade = dollar price of exports * quantity of exports - dollar price of imports * quantity of imports A currency depreciation affects the items on the right side of this equation as follows: First, the quantity of exports increases and the quantity of imports decreases; both have a positive effect
|
on the balance of trade (lowering the trade deficit or raising the trade surplus). Second, the dollar price of exports is not likely to change very much, at least not initially. The dollar price of exports changes when the U.S. price level changes, but the initial effect of a depreciation on the domestic price level is not likely to be large. Third, the dollar price of imports increases. Imports into the United States are more expensive because $1 U.S. buys fewer yen, euros, and so on, than before. An increase in the dollar price of imports has a negative effect on the balance of trade. The following is an example to clarify this last point: The dollar price of a Japanese car that costs 1,200,000 yen rises from $10,000 to $12,000 when the exchange rate moves from 5For this reason, some countries are tempted at times to intervene in foreign exchange markets, depreciate their currencies, and stimulate their economies. If all countries attempted to lower the value of their currencies simultaneously, there would be no gain in income for any of them. Although the exchange rate system at the time was different, such a situation actually occurred during the early years of the Great Depression. Many countries practiced so-called beggar-thy-neighbor policies of competitive devaluations in a desperate attempt to maintain export sales and employment. J-curve effect Following a currency depreciation, a country’s balance of trade may get worse before it gets better. The graph showing this effect is shaped like the letter J, hence the name J-curve effect. M19_CASE3826_13_GE_C19.indd 399 17/04/19 4:26 AM 400 PART V The World Economy ▸ FIGURE 19.7 The Effect of a Depreciation on the Balance of Trade (the J Curve) Initially, a depreciation of a country’s currency may worsen its balance of trade. The negative effect on the price of imports may initially dominate the positive effects of an increase in exports and a decrease in imports Quarters after the beginning of the depreciation 8 7 MyLab Economics Concept Check 120 yen per dollar to 100 yen per dollar. After the currency depreciation, the United States ends up spending more (in dollars) for the Japanese car than it did before. Of course, the United States will end up buying fewer Japanese cars than it did before. Does the number of cars drop enough so that the quantity effect is bigger than the price effect
|
or vice versa? Does the value of imports increase or decrease? The net effect of a depreciation on the balance of trade could go either way. The depreciation stimulates exports and cuts back imports, but it also increases the dollar price of imports. It seems that the negative effect dominates initially. The impact of a depreciation on the price of imports is generally felt quickly, while it takes time for export and import quantities to respond to price changes. In the short run, the value of imports increases more than the value of exports, so the balance of trade worsens. The initial effect is likely to be negative, but after exports and imports have had time to respond, the net effect turns positive. The more elastic the demand for exports and imports is, the larger the eventual improvement in the balance of trade will be. Exchange Rates and Prices The depreciation of a country’s currency tends to increase its price level. There are two reasons for this effect. First, when a country’s currency is less expensive, its products are more competitive on world markets, so exports rise. In addition, domestic buyers tend to substitute domestic products for the now-more-expensive imports. This means that planned aggregate expenditure on domestically produced goods and services rises and that the aggregate demand curve shifts to the right. The result is a higher price level, a higher output, or both. (You may want to draw an AS/AD diagram to verify this outcome.) If the economy is close to capacity, the result is likely to be higher prices. Second, a depreciation makes imported inputs more expensive. If costs increase, the aggregate supply curve shifts to the left. If aggregate demand remains unchanged, the result is an increase in the price level. Monetary Policy with Flexible Exchange Rates Let us now put everything in this chapter together and consider what happens when monetary policy is used first to stimulate the economy and then to contract the economy in an open economy with flexible exchange rates. Suppose the economy is below full employment and the Federal Reserve (Fed) lowers the interest rate. The lower interest rate stimulates planned investment spending and consumption spending. Output thus increases, but there are additional effects: (1) The lower interest rate has an impact in the foreign exchange market. A lower interest rate means a lower demand for U.S. securities by foreigners, so the demand for dollars drops. (2) U.S. investment managers will be more likely to buy foreign securities (which are now paying relatively higher interest rates), so the supply of dollars rises. Both events push down the
|
value of the dollar. A cheaper dollar is a good thing if the goal of the Fed is to stimulate the domestic economy because a cheaper dollar means more U.S. exports and fewer imports. If consumers substitute U.S.-made goods for imports, both the added exports and the decrease in imports mean more spending on domestic products, so the multiplier actually increases. Flexible exchange rates thus help the Fed in its goal to stimulate the economy. Now suppose inflation is a problem and the Fed raises the interest rate. Here again, floating exchange rates help. The higher interest rate lowers planned investment and consumption M19_CASE3826_13_GE_C19.indd 400 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 401 spending, reducing output and lowering the price level. The higher interest rate also attracts foreign buyers into U.S. financial markets, driving up the value of the dollar, which reduces the price of imports. The reduction in the price of imports causes a shift of the aggregate supply curve to the right, which helps fight inflation, which is what the Fed wants to do. Flexible exchange rates thus help the Fed in its goal to fight inflation. Fiscal Policy with Flexible Exchange Rates Although we have just seen that flexible exchange rates help the Fed achieve its goals, the opposite is the case for the fiscal authorities in normal times when there is no zero lower interest rate bound and the Fed is following the Fed rule. Say that the administration and Congress want to stimulate the economy, and they increase government spending to do this. This increases output in the usual way (shift of the AD curve to the right). This usual way means that the interest rate is also higher (from the Fed rule because output and the price level are higher). The higher interest rate attracts foreign investment and leads to an appreciation of the dollar. An appreciation, other things being equal, increases imports and decreases exports, which has a negative effect on output. The increase in output is thus less than it would have been had there been no appreciation. The appreciation also leads to a decrease in import prices, which shifts the AS curve to the right, thus decreasing the price level, other things being equal. Although the price level is lower than otherwise, output, which was the main target of the administration’s policy in our example, is lower, all else equal. Flexible exchange rates thus makes the task of the fiscal authorities in their goal to stimulate the economy
|
more difficult. Flexible exchange rates also hurt the fiscal authorities if they want to contract the economy to fight inflation. Suppose we decrease government spending to try to reduce inflation. This shifts the AD curve to the left, which decreases output and the price level. The interest rate is also lower (from the Fed rule because output and the price level are lower), which leads to a depreciation of the dollar. The depreciation, other things being equal, decreases imports and increases exports, which has a positive effect on output. However, the depreciation also leads to an increase in import prices, which shifts the AS curve to the left, thus increasing the price level, other things being equal. Although output is higher than otherwise, inflation, which was our target, is higher than it would have been in a closed economy other things being equal. So flexible exchange rates also hurt the fiscal authorities in their goal to fight inflation. Note that the appreciation or depreciation of the currency occurs because of the Fed rule. If the Fed does not change the interest rate in response to the fiscal policy change, either because there is a zero lower bound or because it just doesn’t want to, there is no appreciation or depreciation and thus no offset to what the fiscal authorities are trying to do from the existence of flexible exchange rates. Monetary Policy with Fixed Exchange Rates Although most major countries in the world today have a flexible exchange rate (counting for this purpose the eurozone countries as one country), it is interesting to ask what role monetary policy can play when a country has a fixed exchange rate. The answer is, no role. Suppose a country fixes or “pegs” its exchange rate to the value of the dollar? In fact a number of countries do this, including countries like Hong Kong and Singapore, who are heavily reliant on their financial sectors. When a country decides to peg its exchange rate to another currency, say the U.S. dollar, it gives up its power to change its interest rate. Why? Consider a monetary authority of a pegged country that wants to lower its interest rate to stimulate the economy. The problem is that with its interest rate lower than rates abroad, people in the country will be induced to move their capital abroad to earn the higher interest rates. In other words, there will be an outflow of capital. Normally, this outflow would cause the country’s currency to depreciate, but with a pegged rate, this won’t happen. To keep the exchange rate from depreciating, the country�
|
�s monetary authority will be forced to buy the domestic currency outflow by selling its foreign reserves. Eventually the monetary authority will run out of foreign reserves and thus be unable to support the pegged exchange rate. It is thus not feasible for the country to change its interest rate and keep its exchange rate unchanged. A commitment to peg is thus a commitment to give up one’s independent monetary policy. When the various European countries moved in 1999 to a common currency, the euro, each country gave up its monetary policy. Monetary policy is decided for all of the eurozone M19_CASE3826_13_GE_C19.indd 401 17/04/19 4:26 AM 402 PART V The World Economy countries by the European Central Bank (ECB). The Bank of Italy, for example, no longer has any influence over Italian interest rates. Interest rates are influenced by the ECB. This is the price Italy paid for giving up the lira. The one case in which a country can change its interest rate and keep its exchange rate fixed is if it imposes capital controls. Imposing capital controls means that the country limits or prevents people from buying or selling its currency in the foreign exchange markets. A citizen of the country may be prevented, for example, from using the country’s currency to buy dollars. The problem with capital controls is that they are hard to enforce, especially for large countries and for long periods of time. An Interdependent World Economy The increasing interdependence of countries in the world economy has made the problems facing policymakers more difficult. We used to be able to think of the United States as a relatively self-sufficient region. Forty years ago economic events outside U.S. borders had relatively little effect on its economy. This situation is no longer true. The events of the past four decades have taught us that the performance of the U.S. economy is heavily dependent on events outside U.S. borders. This chapter and the previous chapter have provided only the bare bones of open-economy macroeconomics. If you continue your study of economics, more will be added to the basic story we have presented. The next chapter concludes with a discussion of the problems of developing countries. S U M M A R Y 1. The main difference between an international transaction and a domestic transaction concerns currency exchange: When people in different countries buy from and sell to each other, an exchange of currencies must also take place. 2. The exchange rate is the price of one country’s currency in
|
terms of another country’s currency. 19.1 THE BALANCE OF PAYMENTS p. 386 3. Foreign exchange is all currencies other than the domestic currency of a given country. The record of a nation’s transactions in goods, services, and assets with the rest of the world is its balance of payments. The balance of payments is also the record of a country’s sources (supply) and uses (demand) of foreign exchange. 19.2 EQUILIBRIUM OUTPUT (INCOME) IN AN OPEN ECONOMY p. 389 4. In an open economy, some income is spent on foreign produced goods instead of domestically produced goods. To measure planned domestic aggregate expenditure in an open economy, we add total exports but subtract total imports: C + I + G + EX - IM. The open economy is in equilibrium when domestic aggregate output (income) (Y) equals planned domestic aggregate expenditure. 5. In an open economy, the multiplier equals 1 [1 - (MPC - MPM)], > where MPC is the marginal propensity to consume and MPM is the marginal propensity to import. The marginal propensity to import is the change in imports caused by a $1 change in income. 6. In addition to income, other factors that affect the level of imports are the after-tax real wage rate, after-tax nonlabor income, interest rates, and relative prices of domestically produced and foreign-produced goods. The demand for exports is determined by economic activity in the rest of the world and by relative prices. 7. An increase in U.S. economic activity leads to a worldwide increase in economic activity, which then “feeds back” to the United States. An increase in U.S. imports increases other countries’ exports, which stimulates economies and increases their imports, which increases U.S. exports, which stimulates the U.S. economy and increases its imports, and so on. This is the trade feedback effect. 8. Export prices of other countries affect U.S. import prices. The general rate of inflation abroad is likely to affect U.S. import prices. If the inflation rate abroad is high, U.S. import prices are likely to rise. 9. One country’s exports are another country’s imports, thus an increase in export prices increases other countries’ import prices. An increase in other countries’ import prices leads to an increase in their domestic prices—and their export
|
prices. In short, export prices affect import prices and vice versa. This price feedback effect shows that inflation is “exportable”; an increase in the price level in one country can drive up prices in other countries, making inflation in the first country worse. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 402 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 403 19.3 THE OPEN ECONOMY WITH FLEXIBLE EXCHANGE RATES p. 393 10. The equilibrium exchange rate occurs when the quantity demanded of a foreign currency in the foreign exchange market equals the quantity of that currency supplied in the foreign exchange market. 11. Depreciation of a currency occurs when a nation’s currency falls in value relative to another country’s currency. Appreciation of a currency occurs when a nation’s currency rises in value relative to another country’s currency. 12. According to the law of one price, if the costs of transportation are small, the price of the same good in different countries should be roughly the same. The theory that exchange rates are set so that the price of similar goods in different countries is the same is known as the purchasing-power-parity theory. In practice, transportation costs are significant for many goods, and the law of one price does not hold for these goods. 13. A high rate of inflation in one country relative to another country puts pressure on the exchange rate between the two countries. There is a general tendency for the currencies of relatively high-inflation countries to depreciate. 14. A depreciation of the dollar tends to increase U.S. GDP by making U.S. exports cheaper (hence, more competitive abroad) and by making U.S. imports more expensive (encouraging consumers to switch to domestically produced goods and services). 15. The effect of a depreciation of a nation’s currency on its balance of trade is unclear. In the short run, a currency depreciation may increase the balance-of-trade deficit because it raises the price of imports. Although this price increase causes a decrease in the quantity of imports demanded, the impact of a depreciation on the price of imports is generally felt
|
quickly, but it takes time for export and import quantities to respond to price changes. The initial effect is likely to be negative, but after exports and imports have had time to respond, the net effect turns positive. The tendency for the balance-of-trade deficit to widen and then to decrease as the result of a currency depreciation is known as the J-curve effect. 16. The depreciation of a country’s currency tends to raise its price level for two reasons. First, a currency depreciation increases planned aggregate expenditure, an effect that shifts the aggregate demand curve to the right. If the economy is close to capacity, the result is likely to be higher prices. Second, a depreciation makes imported inputs more expensive. If costs increase, the aggregate supply curve shifts to the left. If aggregate demand remains unchanged, the result is an increase in the price level. 17. When exchange rates are flexible, a U.S. expansionary monetary policy decreases the interest rate and stimulates planned investment and consumption spending. The lower interest rate leads to a lower demand for U.S. securities by foreigners and a higher demand for foreign securities by U.S. investment-fund managers. As a result, the dollar depreciates. A U.S. contractionary monetary policy appreciates the dollar. 18. Flexible exchange rates do not always work to the advantage of policy makers. An expansionary fiscal policy can appreciate the dollar and work to reduce the multiplier appreciation of a currency, p. 396 balance of payments, p. 386 balance of trade, p. 387 balance on current account, p. 387 depreciation of a currency, p. 396 exchange rate, p. 386 floating, or market-determined, exchange rates, p. 394 foreign exchange, p. 386 J-curve effect, p. 399 law of one price, p. 397 marginal propensity to import (MPM), p. 390 net exports of goods and services (EX - IM), p. 390 price feedback effect, p. 393 purchasing-power-parity theory, p. 397 trade deficit, p. 387 trade feedback effect, p. 392 P R O B L E M S All problems are available on MyLab Economics. Equations: Planned aggregate expenditure in an open economy: AE K C + I + G + EX - IM, p. 390 Open-economy multiplier = 1 1 - (MPC - MPM), p. 391 19.1 THE BALANCE OF PAYMENTS LEAR
|
NING OBJECTIVE: Explain how the balance of payments is calculated. 1.1 Obtain a recent issue of The Economist. Turn to the section titled “Economic and financial indicators.” Look at the table titled “Trade, exchange rates, budget balances, and interest rates.” Which country had the largest trade deficit over the last year and during the last month? Which country had the largest trade surplus over the last year and during the last month? How does the current account deficit/surplus compare to the overall trade balance? How can you explain the difference? 1.2 What effect will each of the following events have on the current account balance if the exchange rate is fixed? If the exchange rate is floating? a. The Indian government raises taxes and income falls. b. The Chinese inflation rate increases, and prices in China rise faster than those in countries with which China trades. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 403 17/04/19 4:26 AM 404 PART V The World Economy c. India adopts a contractionary monetary policy. Interest rates rise (and are now higher than those in other countries) and income falls. d. The textile companies’ “Buy Japanese” campaign is successful, and Japanese consumers switch from purchasing imported products to buying products made in Japan. 1.3 [Related to the Economics in Practice on p. 389] The United States is the second largest oil importer in the world, importing 7.2 million barrels of crude oil per day as in April 2015. Go to www.inflationdata.com and look up crude oil prices for the past five years; then go to www.bea.gov to look up the U.S. net international investment position (NIIP) for the past five years. Does there appear to be a relationship between the price of crude oil and U.S. NIIP? Briefly explain the result of your findings. 19.2 EQUILIBRIUM OUTPUT (INCOME) IN AN OPEN ECONOMY LEARNING OBJECTIVE: Discuss how equilibrium output is determined in an open economy, and describe the trade feedback effect and the price feedback effect. 2.1 The exchange rate between the U.S. dollar
|
and the euro is floating freely—both governments do not intervene in the market for each currency. Suppose the president of the United States decides to place tariffs on certain imports from Europe. He has also argued in many places that the dollar is too strong visá-vis other currencies and it should be weakened so that the United States regains some of its competitive advantage. a. How will consumption change in the United States? b. What will happen if the income rises in the United States? c. How will a decline in the export of European goods im- pact the dollar-euro exchange rate? d. Consider the effects of a tariff on European imports. How will the European Union respond to U.S. tariffs? 2.2 You are given the following model that describes the economy of Hypothetica. (1) Consumption function: C = 80 + 0.75Yd (2) Planned investment: I = 49 (3) Government spending: G = 60 (4) Exports: EX = 20 (5) Imports: IM = 0.05Yd (6) Disposable income: Yd = Y - T (7) Taxes: T = 20 (8) Planned aggregate expenditure: AE = C = I + G + EX - IM (9) Definition of equilibrium income: Y = AE a. What is equilibrium income in Hypothetica? What is the government deficit? What is the current account balance? b. If government spending is increased to G = 75, what happens to equilibrium income? Explain using the government spending multiplier. What happens to imports? c. Now suppose the amount of imports is limited to IM = 25 by a quota on imports. If government spending is again increased from 60 to 75, what happens to equilibrium income? Explain why the same increase in G has a bigger effect on income in the second case. What is it about the presence of imports that changes the value of the multiplier? d. If exports are fixed at EX = 20, what must income be to ensure a current account balance of zero? (Hint: Imports depend on income, so what must income be for imports to be equal to exports?) By how much must we cut government spending to balance the current account? (Hint: Use your answer to the first part of this question to determine how much of a decrease in income is needed. Then use the multiplier to calculate the decrease in G needed to reduce income by that amount.) 19.3 THE OPEN ECONOMY
|
WITH FLEXIBLE EXCHANGE RATES LEARNING OBJECTIVE: Discuss factors that affect exchange rates in an open economy with a floating system. 3.1 On January 3, 2014, the euro was trading at $1.36658, while on August 20, 2018, the euro was worth $1.15. What reasons can you give for the dollar strengthening vis-á-vis the euro? 3.2 Suppose the following graph shows what prevailed on the foreign exchange market in 2018 with floating exchange rates. a. Name three phenomena that might shift the demand curve to the right. b. Which, if any, of these three phenomena might cause a simultaneous shift of the supply curve to the left? c. What effects might each of the three phenomena have on the balance of trade if the exchange rate floats1.60 S D 0 Quantity of pounds 3.3 Suppose the exchange rate between the Danish krone and the U.S. dollar is 7 DKK = $1 and the exchange rate between the Chilean peso and the U.S. dollar is 650 CLP = $1. a. Express both of these exchange rates in terms of dollars per unit of the foreign currency. b. What should the exchange rate be between the Danish krone and the Chilean peso? Express the exchange rate in terms of one krone and in terms of one peso. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 404 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 405 c. Suppose the exchange rate between the krone and the dollar changes to 5 DKK = $1 and the exchange rate between the peso and the dollar changes to 700 CLP = $1. For each of the three currencies, explain whether the currency has appreciated or depreciated against the other two currencies. 3.4 Suppose the exchange rate between the British pound and the U.S. dollar is £1 = $1.50. a. Draw a graph showing the demand and supply of pounds for dollars. b. If the Bank of England implements a contractionary monetary policy, explain what will happen to the exchange rate between the pound and the dollar and show this on a graph. Has the
|
dollar appreciated or depreciated relative to the pound? Explain. c. If the U.S. government implements an expansionary fiscal policy, explain what will happen to the exchange rate between the pound and the dollar and show this on a graph. Has the dollar appreciated or depreciated relative to the pound? Explain. 3.5 Canada is the second-largest trading partner for the United States (just recently surpassed by China). In 2017, U.S. exports to Canada were more than $340 billion and imports from Canada totaled more than $332 billion. On January 1, 2017, the exchange rate between the Canadian dollar and the U.S. dollar was 1.34 Canadian dollars = 1 U.S. dollar. On January 1, 2018, the exchange rate was 1.26 Canadian dollars = 1 U.S. dollar. Explain how this change in exchange rates could impact U.S. consumers and firms? 3.6 The exchange rate between the U.S. dollar and the British pound is a floating rate, with no government intervention. If a large trade deficit with Great Britain prompts the United States to impose quotas on certain British imports, resulting in a reduction in the quantity of these imports, what will happen to the dollar–pound exchange rate? Why? (Hint: There is an excess supply of pounds, or an excess demand for dollars.) What effects will the change in the value of each currency have on employment and output in the United States? What about the balance of trade? (Ignore complications such as the J curve.) 3.7 Do an Internet search and look up historical exchange rates. Find the recent exchange rates between the Indian rupee and the Chinese yuan and between the U.S. dollar and the Emirati dirham. Compare them with exchange rates a year ago. Go to the website of Ministry of Commerce and Industry, India, to find the latest value of Indian exports, imports, and trade balance and compare with estimates of the previous year. Did these values increase or decrease during that year? Explain how changes in the exchange rates may have had an impact on the changes in Indian exports, imports, and the trade balance. Discuss if you witness any deviation from the theories studied. 3.8 The data in the following table represents price level changes and interest rate changes over a one-year period for three countries: Astoria, Borgia, and Calistoga. Based on the data, explain what is likely to happen to the exchange rate for Astorian asters relative
|
to the other two countries’ currencies over that one-year period. Use supply and demand graphs to support your answer, with prices listed as asters per borg and asters per cali, and quantities representing borgs and calis. Country/ Currency Price Index January 1, 2018 Price Index January 1, 2019 Interest Rate January 1, 2018 Interest Rate January 1, 2019 Astoria/aster Borgia/borg Calistoga/cali 100 120 150 110 132 168 4 percent 4 percent 4 percent 6 percent 8 percent 6 percent QUESTION 1 Tania is a Costa Rican graduate student at the University of Florida. She earns a small salary as a graduate student, and she sends a portion of that income to her family in Costa Rica as a remittance. Would this transfer appear as a positive or negative entry in the U.S. Current Account? And in the Costa Rican Current Account? QUESTION 2 During 2015 and 2016, the Mexican Peso depreciated substantially in value, and many Mexican commentators lamented a tragic occurrence and blamed thenPresident Enrique Peña Nieto. Why might this depreciation of the Mexican Peso have been a good thing? CHAPTER 19 APPENDIX: World Monetary Systems since 1900 Since the beginning of the twentieth century, the world has operated under a number of different monetary systems. This Appendix provides a brief history of each and a description of how they worked. LEARNING OBJECTIVE Explain what the Bretton Woods system is. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 405 17/04/19 4:26 AM 406 PART V The World Economy The Gold Standard MyLab Economics Concept Check The gold standard was the major system of exchange rate determination before 1914. All currencies were priced in terms of gold—an ounce of gold was worth so much in each currency. When all currencies exchanged at fixed ratios to gold, exchange rates could be determined easily. For instance, one ounce of gold was worth $20 U.S.; that same ounce of gold exchanged for £4 (British pounds). Because $20 and £4 were each worth one ounce of gold, the exchange rate between dollars and pounds was $20/£4, or $5 to £1. For the gold standard to be effective, it had to be backed up by the country�
|
�s willingness to buy and sell gold at the determined price. As long as countries maintain their currencies at a fixed value in terms of gold and as long as each country is willing to buy and sell gold, exchange rates are fixed. If at the given exchange rate the number of U.S. citizens who want to buy things produced in Great Britain is equal to the number of British citizens who want to buy things produced in the United States, the currencies of the two countries will simply be exchanged. What if U.S. citizens suddenly decide they want to drink imported Scotch instead of domestic bourbon? If the British do not have an increased desire for U.S. goods, they will still accept U.S. dollars because those dollars can be redeemed in gold. This gold can then be immediately turned into pounds. As long as a country’s overall balance of payments remained in balance, no gold would enter or leave the country and the economy would be in equilibrium. If U.S. citizens bought more from the British than the British bought from the United States, however, the U.S. balance of payments would be in deficit and the U.S. stock of gold would begin to fall. Conversely, Britain would start to accumulate gold because it would be exporting more than it spent on imports. Under the gold standard, gold was a big determinant of the money supply.6 An inflow of gold into a country caused that country’s money supply to expand, and an outflow of gold caused that country’s money supply to contract. If gold were flowing from the United States to Great Britain, the British money supply would expand and the U.S. money supply would contract. Now recall from previous chapters the impacts of a change in the money supply. An expanded money supply in Britain will lower British interest rates and stimulate aggregate demand. As a result, aggregate output (income) and the price level in Britain will increase. Higher British prices will discourage U.S. citizens from buying British goods. At the same time, British citizens will have more income and will face relatively lower import prices, causing them to import more from the States. On the other side of the Atlantic, U.S. citizens will face a contracting domestic money supply. This will cause higher interest rates, declining aggregate demand, lower prices, and falling output (income). The effect will be lower demand in the United States for British goods. Thus, changes in relative prices and incomes that resulted from the inflow and outflow of
|
gold would automatically bring trade back into balance. Problems with the Gold Standard MyLab Economics Concept Check Two major problems were associated with the gold standard. First, the gold standard implied that a country had little control over its money supply. The reason, as we have just seen, is that the money stock increased when the overall balance of payments was in surplus (gold inflow) and decreased when the overall balance was in deficit (gold outflow). A country that was experiencing a balance-of-payments deficit could correct the problem only by the painful process of allowing its money supply to contract. This contraction brought on a slump in economic activity, a slump that would eventually restore balance-of-payments equilibrium, but only after reductions in income and employment. Countries could (and often did) act to protect their gold reserves, and this precautionary step prevented the adjustment mechanism from correcting the deficit. Making the money supply depend on the amount of gold available had another disadvantage. When major new gold fields were discovered (as in California in 1849 and South Africa in 6In the days when currencies were tied to gold, changes in the amount of gold influenced the supply of money in two ways. A change in the quantity of gold coins in circulation had a direct effect on the supply of money; indirectly, gold served as a backing for paper currency. A decrease in the central bank’s gold holdings meant a decline in the amount of paper money that could be supported. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 406 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 407 1886), the world’s supply of gold (and therefore of money) increased. The price level rose and income increased. When no new gold was discovered, the supply of money remained unchanged and prices and income tended to fall. When President Reagan took office in 1981, he established a commission to consider returning the nation to the gold standard. The final commission report recommended against such a move. An important part of the reasoning behind this recommendation was that the gold standard puts enormous economic power in the hands of gold-producing nations. Fixed Exchange Rates and the Bretton Woods System MyLab Economics Concept Check As World War II drew to a close, a
|
group of economists from the United States and Europe met to formulate a new set of rules for exchange rate determination that they hoped would avoid the difficulties of the gold standard. The rules they designed became known as the Bretton Woods system, after the town in New Hampshire where the delegates met. The Bretton Woods system was based on two (not necessarily compatible) premises. First, countries were to maintain fixed exchange rates with one another. Instead of pegging their currencies directly to gold, however, currencies were fixed in terms of the U.S. dollar, which was fixed in value at $35 per ounce of gold. The British pound, for instance, was fixed at roughly $2.40, so that an ounce of gold was worth approximately £14.6. As we shall see, the pure system of fixed exchange rates would work in a manner very similar to the pre-1914 gold standard. The second aspect of the Bretton Woods system added a new wrinkle to the operation of the international economy. Countries experiencing a “fundamental disequilibrium” in their balance of payments were allowed to change their exchange rates. (The term fundamental disequilibrium was necessarily vague, but it came to be interpreted as a large and persistent current account deficit.) Exchange rates were not really fixed under the Bretton Woods system; they were, as someone remarked, only “fixed until further notice.” The point of allowing countries with serious current account problems to alter the value of their currency was to avoid the harsh recessions that the operation of the gold standard would have produced under these circumstances. However, the experience of the European economies in the years between World War I and World War II suggested that it might not be a good idea to give countries complete freedom to change their exchange rates whenever they wanted. During the Great Depression, many countries undertook so-called competitive devaluations to protect domestic output and employment. That is, countries would try to encourage exports—a source of output growth and employment—by attempting to set as low an exchange rate as possible, thereby making their exports competitive with foreign-produced goods. Unfortunately, such policies had a built-in flaw. A devaluation of the pound against the French franc might help encourage British exports to France, but if those additional British exports cut into French output and employment, France would likely respond by devaluing the franc against the pound, a move that, of course, would undo the effects of the pound’s initial devaluation. To solve this exchange rate rivalry,
|
the Bretton Woods agreement created the International Monetary Fund (IMF). Its job was to assist countries experiencing temporary current account problems.7 It was also supposed to certify that a “fundamental disequilibrium” existed before a country was allowed to change its exchange rate. The IMF was like an international economic traffic cop whose job was to ensure that all countries were playing the game according to the agreed-to rules and to provide emergency assistance where needed. “Pure” Fixed Exchange Rates MyLab Economics Concept Check Under a pure fixed exchange rate system, governments set a particular fixed rate at which their currencies will exchange for one another and then commit themselves to maintaining that rate. A true fixed exchange rate system is like the gold standard in that exchange rates are supposed 7The idea was that the IMF would make short-term loans to a country with a current account deficit. The loans would enable the country to correct the current account problem gradually, without bringing on a deep recession, running out of foreign exchange reserves, or devaluing the currency. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 407 17/04/19 4:26 AM 408 PART V The World Economy ▸ FIGURE 19A.1 Government Intervention in the Foreign Exchange Market If the price of Australian dollars were set in a completely unfettered market, one Australian dollar would cost 0.96 U.S. dollars when demand is D0 and 0.90 when demand is D1. If the government has committed to keeping the value at 0.96, it must buy up the excess supply of Australian dollars (Qs - Qd). 0.96 $0.90 S Excess supply of Australian dollars 5 intervention D0 D1 MyLab Economics Concept Check Quantity of Australian dollars 0 Qd Qs to stay the same forever. Because currencies are no longer backed by gold, they have no fixed, or standard, value relative to one another. There is, therefore, no automatic mechanism to keep exchange rates aligned with each other, as with the gold standard. The result is that under a pure fixed exchange rate system, governments must at times intervene in the foreign exchange market to keep currencies aligned at their established values. Economists define government intervention in the foreign exchange market as the buying or selling of foreign exchange for the purpose
|
of manipulating the exchange rate. What kind of intervention is likely to occur under a fixed exchange rate system, and how does it work? We can see how intervention works by looking at Figure 19.A.1. Initially, the market for Australian dollars is in equilibrium. At the fixed exchange rate of 0.96, the supply of dollars is exactly equal to the demand for dollars. No government intervention is necessary to maintain the exchange rate at this level. Now suppose Australian wines are found to be contaminated with antifreeze and U.S. citizens switch to California wines. This substitution away from the Australian product shifts the U.S. demand curve for Australian dollars to the left: The United States demands fewer Australian dollars at every exchange rate (cost of an Australian dollar) because it is purchasing less from Australia than it did before. If the price of Australian dollars were set in a completely unfettered market, the shift in the demand curve would lead to a fall in the price of Australian dollars, just the way the price of wheat would fall if there was an excess supply of wheat. Remember, the Australian and U.S. governments have committed themselves to maintaining the rate at 0.96. To do so, either the U.S. government or the Australian government (or both) must buy up the excess supply of Australian dollars to keep its price from falling. In essence, the fixed exchange rate policy commits governments to making up any difference between the supply of a currency and the demand so as to keep the price of the currency (exchange rate) at the desired level. The government promises to act as the supplier (or demander) of last resort, who will ensure that the amount of foreign exchange demanded by the private sector will equal the supply at the fixed price. Problems with the Bretton Woods System MyLab Economics Concept Check As it developed after the end of World War II, the system of more-or-less fixed exchange rates had some flaws that led to its abandonment in 1971. First, there was a basic asymmetry built into the rules of international finance. Countries experiencing large and persistent current account deficits—what the Bretton Woods agreements termed “fundamental disequilibria”—were obliged to devalue their currencies and/or take measures to cut their deficits by contracting their economies. Both of these alternatives were MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are
|
marked with. M19_CASE3826_13_GE_C19.indd 408 17/04/19 4:26 AM CHAPTER 19 Open-Economy Macroeconomics: The Balance of Payments and Exchange Rates 409 unpleasant because devaluation meant rising prices and contraction meant rising unemployment. However, a country with a current account deficit had no choice because it was losing stock of foreign exchange reserves. When its stock of foreign currencies became exhausted, it had to change its exchange rate because further intervention (selling off some of its foreign exchange reserves) became impossible. Countries experiencing current account surpluses were in a different position because they were gaining foreign exchange reserves. Although these countries were supposed to stimulate their economies and/or revalue their currencies to restore balance to their current account, they were not obliged to do so. They could easily maintain their fixed exchange rate by buying up any excess supply of foreign exchange with their own currency, of which they had plentiful supply. In practice, this meant that some countries—especially Germany and Japan—tended to run large and chronic current account surpluses and were under no compulsion to take steps to correct the problem. The U.S. economy, stimulated by expenditures on the Vietnam War, experienced a large and prolonged current account deficit (capital outflow) in the 1960s, which was the counterpart of these surpluses. The United States was, however, in a unique position under the Bretton Woods system. The value of gold was fixed in terms of the U.S. dollar at $35 per ounce of gold. Other countries fixed their exchange rates in terms of U.S. dollars (and therefore only indirectly in terms of gold). Consequently, the United States could never accomplish anything by devaluing its currency in terms of gold. If the dollar was devalued from $35 to $40 per ounce of gold, the yen, pegged at 200 yen per dollar, would move in parallel with the dollar (from 7,000 yen per ounce of gold to 8,000 yen per ounce), with the dollar–yen exchange rate unaffected. To correct its current account deficits vis-à-vis Japan and Germany, it would be necessary for those two countries to adjust their currencies’ exchange rates with the dollar. These countries were reluctant to do so for a variety of reasons. As a result, the U.S. current account was chronically in deficit throughout the late 1960s. A second flaw in the Bretton Woods system was that it permitted devaluations only when
|
a country had a “chronic” current account deficit and was in danger of running out of foreign exchange reserves. This meant that devaluations could often be predicted quite far in advance, and they usually had to be rather large if they were to correct any serious current account problem. The situation made it tempting for speculators to “attack” the currencies of countries with current account deficits. Problems such as these eventually led the United States to abandon the Bretton Woods rules in 1971. The U.S. government refused to continue pegging the value of the dollar in terms of gold. Thus, the prices of all currencies were free to find their own levels. The alternative to fixed exchange rates is a system that allows exchange rates to move freely or flexibly in response to market forces. Two types of flexible exchange rate systems are usually distinguished. In a freely floating system, governments do not intervene at all in the foreign exchange market.8 They do not buy or sell currencies with the aim of manipulating the rates. In a managed floating system, governments intervene if markets are becoming “disorderly”—fluctuating more than a government believes is desirable. Governments may also intervene if they think a currency is increasing or decreasing too much in value even though the day-to-day fluctuations may be small. Since the demise of the Bretton Woods system in 1971, the world’s exchange rate system can be described as “managed floating.” One of the important features of this system has been times of large fluctuations in exchange rates. For example, the yen–dollar rate went from 347 in 1971 to 210 in 1978, to 125 in 1988, and to 80 in 1995. Those are very large changes, changes that have important effects on the international economy, some of which we have covered in this text. 8However, governments may from time to time buy or sell foreign exchange for their own needs (instead of influencing the exchange rate). For example, the U.S. government might need British pounds to buy land for a U.S. embassy building in London. For our purposes, we ignore this behavior because it is not “intervention” in the strict sense of the word. MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 409
|
17/04/19 4:26 AM 410 PART V The World Economy. The gold standard was the major system of exchange rate determination before 1914. All currencies were priced in terms of gold. Difficulties with the gold standard led to the Bretton Woods agreement following World War II. Under this system, countries maintained fixed exchange rates with one another and fixed the value of their currencies in terms of the U.S. dollar. Countries experiencing a “fundamental disequilibrium” in their current accounts were permitted to change their exchange rates. 2. The Bretton Woods system was abandoned in 1971. Since then, the world’s exchange rate system has been one of managed floating rates. Under this system, governments intervene if foreign exchange markets are fluctuating more than the government thinks desirable All problems are available on MyLab Economics. CHAPTER 19 APPENDIX: WORLD MONETARY SYSTEMS SINCE 1900 LEARNING OBJECTIVE: Explain what the Bretton Woods system is. 1A.1 The currency of Atlantis is the wimp. In 2018, Atlantis developed a balance-of-payments deficit with the United States as a result of an unanticipated decrease in exports; U.S. citizens cut back on the purchase of Atlantean goods. Assume Atlantis is operating under a system of fixed exchange rates. a. How does the drop in exports affect the market for wimps? Identify the deficit graphically. b. How must the government of Atlantis act (in the short run) to maintain the value of the wimp? c. If Atlantis had originally been operating at full employment (potential GDP), what impact would those events have had on its economy? Explain your answer. d. The chief economist of Atlantis suggests an expansionary monetary policy to restore full employment; the Secretary of Commerce suggests a tax cut (expansionary fiscal policy). Given the fixed exchange rate system, describe the effects of these two policy options on Atlantis’s current account. e. How would your answers to a, b, and c change if the two countries operated under a floating rate system? MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M19_CASE3826_13_GE_C19.indd 410 17/04/19 4:26 AM Economic Growth in Developing Economies 20 CHAPTER OUTLINE
|
AND LEARNI NG OBJECTIV ES 20.1 Life in the Developing Nations: Population and Poverty p. 412 Discuss the characteristics of developing nations. 20.2 Economic Development: Sources and Strategies p. 413 Describe the sources of economic development. 20.3 Development Interventions p. 422 Discuss the intervention methods used by development economists. 411 In 2000 all 189 member states of the United Nations (UN) agreed to work toward achieving a set of eight Millennium Development Goals (MDG) for the developing world by 2015. Goals ranged from eradicating hunger and achieving universal primary education to reducing child and maternal mortality to fostering gender equality and environmental sustainability. After 15 years, in 2015, the UN reported considerable progress on most of its goals. Child and maternal mortality were both cut in half over the 15-year period; deaths from both HIV and malaria dramatically decreased and primary school attendance increased. Nevertheless, in its report on this progress, the UN concluded: “The poorest and most vulnerable people are being left behind.”1 What accounts for the persistence of underdevelopment across the world and what can be done about it? We will begin our discussion in this chapter with a look at some data comparing the developing and developed world. One of the hallmarks of the UN effort was its focus on the importance of collecting and analyzing data. With this context, we turn to look at strategies for economic development generally and then look at evidence on some specific interventions in the developing world, largely focused on the poorest households. As part of this discussion we will touch on some methodological questions current in economics about how best to determine whether particular policy interventions work or do not work. 1UN, Millennium Development Goals 2015 Report. M20_CASE3826_13_GE_C20.indd 411 17/04/19 5:34 PM 412 PART V The World Economy 20.1 LEARNING OBJECTIVE Discuss the characteristics of developing nations. Global South Developing Nations in Asia, Africa, and Latin America. Life in the Developing Nations: Population and Poverty In 2015, the population of the world reached more than 7 billion people. Most of the world’s more than 200 nations belong to the developing world, also known as the Global South, in which about three-fourths of the world’s population lives. In the last decade, rapid economic progress has brought some developing nations closer to developed economies. People living in extreme poverty, earning less than $1.25 per day, declined from 47 percent of
|
the world’s population to 14 percent. Countries such as Argentina and Chile, still considered part of the Global South, have vibrant middle classes. Russia and many countries in the former Soviet bloc have also climbed to middle-income status. China and India, while still experiencing some of the challenges of the Global South, are becoming economic superpowers. At present, China’s gross domestic product (GDP) is second only in the world to the United States. Other parts of the world, most notably parts of Asia and sub-Saharan Africa, lag behind on many of the central dimensions of well-being identified by the UN and others. A central challenge in development economics is to explain why some countries lag and whether successful strategies of the past have lessons for the countries still left behind. Table 20.1 describes the progress of a dozen nations from 1990 to 2013 on two of the measures of human capital targeted by the MDG, child mortality younger than age five and literacy. As we will see in the next section, health and education are two of the lynch pins of economic development. If you think back to our discussion of economic growth in Chapter 16, you will recall the importance of human capital in promoting economic growth. What do the data tell us? The good news is that on both measures progress has been made over the last 25 years. In all countries, developed and developing, child mortality has fallen and literacy has risen. The improvement in child mortality in China is especially notable. But the disparity between Global North and Global South remains high. In 2014 in the sub-Saharan countries, one in 10 children die before they are five. In some of the countries in Africa, including Niger, Chad, and Central African Republic, illiteracy remains high at more than half the adult population. Moreover, even as the Global South increases its primary education levels, the Global North is providing college educations to a larger fraction of their populations, preserving its human capital lead. Although the countries of the developing world exhibit considerable diversity in both their standards of living and their particular experiences of growth, marked differences continue to separate them from the developed nations. The great majority of the population in the Global South lives in rural areas where agricultural work is hard and extremely time-consuming. Productivity (output produced per worker) is low in part because farmers work with little capital. Low productivity means farm output per person is barely sufficient to feed a farmer’s own family. The UN figures indicate that in TABLE 20.1 Comparisons of Child Mortality
|
and Literacy: Selected Countries 1990 and 2013 Country Afghanistan Angola Australia Chad Central African Republic China Denmark Guinea Bissau India Niger Sierra Leone United States 1990: Mortality younger than age 5 2013: Mortality younger than age 5 1990: Literacy rates, ages 15–24 2013 Literacy rates: ages 15–24 179.1 225.9 9.2 214.7 176.9 53.9 8.9 224.8 125.9 327.3 267.7 11.2 97.3 167.4 4.0 147.5 139.2 12.7 3.5 123.9 52.7 104.2 160.6 6.9 Na Na 100.0 17.3 Na 94.3 100.0 Na 61.9 Na Na 100.0 47.0 73.0 100.0 48.9 36.4 99.6 100.0 74.3 81.1 23.5 62.7 100.0 Source: UN, Millennium Development Goal Data, 2015. M20_CASE3826_13_GE_C20.indd 412 17/04/19 5:34 PM What Can We Learn from the Height of Children? CHAPTER 20 Economic Growth in Developing Economies 413 The first of the Millennium Development Goals is to substantially cut the number of households who experience extreme hunger. One of four children younger than five years of age in the world are characterized as stunted, extremely short because of malnutrition. Of these children, one half is in Asia and one third in Africa. For these children poor nutrition in the early years leaves a permanent mark, reflected in life span and earnings. Recent work in economics has focused on the case of stunting in India. India’s stunting rate is among the highest in the world, exceeding even that of the much poorer African nations. Moreover, despite rapid growth in the last decade, little progress has been made in reducing the stunting rate. Seema Jayanchandran from Northwestern and Rohini Pande of Harvard’s Kennedy School examined several large data sets to try to understand why.1 The first clue comes from the pattern of India’s stunting. Looking at the data, Jayannchandran and Pande learn that Indian first born sons are actually taller than their African counterparts. Stunting emerges only for later born children, and the amount of stunting increases with the number of children. Among the most disadvantaged are girls with no older brothers whose parents continue to attempt to
|
produce a son. The patterns that emerge from this study put a spotlight on two of the MDG concerns: hunger and gender equality. The researchers argue that India’s high stunting rate is explicable by the strong son preference of Indian families and the concomitant decision to invest disproportionate family resources in the first born son to insure his survival despite the family’s poverty. CRITICAL THINKING 1. Why might growth in the overall economy not have led to more improvement in the stunting rate? 1Seema Jayachandran and Rohini Pande, “Why are Indian Children so Short?” American Economic Review 2017, 2600–2629. 2015, 836 million people, primarily in the developing world, were severely undernourished. In addition, many developing nations are engaged in civil and external warfare. In recent years there has been more concern with the increased inequality that has come with development in some countries. India is on the World Bank’s list of low-income countries, yet Mumbai, a state capital, is one of the top 10 centers of commerce in the world, home to Bollywood, the world’s largest film industry. China with its rapid growth rates and increased affluence in urban areas still has a large agrarian population that has been mostly left behind by recent growth. Many of the specific interventions we will look at in this chapter are focused on designing strategies to bringing the households at the bottom rung of the income distribution into the mainstream economy of a country. Economic Development: Sources and Strategies Economists have been trying to understand economic growth and development since Adam Smith and David Ricardo in the eighteenth and nineteenth centuries, but the study of development economics as it applies to the developing nations has a much shorter history. The geopolitical struggles that followed World War II brought increased attention to the developing nations and their economic problems. During this period, the new field of development economics asked simply: Why are some nations poor and others rich? If economists could understand the barriers 20.2 LEARNING OBJECTIVE Describe the sources of economic development. M20_CASE3826_13_GE_C20.indd 413 17/04/19 5:34 PM 414 PART V The World Economy to economic growth that prevent nations from developing and the prerequisites that would help them to develop, economists could prescribe strategies for achieving economic advancement. We will see in this discussion that there is lively debate on the question of why some nations are poor and the cor
|
ollary, how can we help countries get out of poverty. Ahijit Banerjee and Esther Duflo, both John Bates Clark award winners and MIT professors, on the other hand, argue in their influential book Poor Economics2 that it is not really possible at this point to answer the question of why some countries are poor and others rich and that the more relevant question is what types of policy interventions help households get out of poverty. We move to that discussion in the last section of this chapter. The Sources of Economic Development MyLab Economics Concept Check Although a general theory of economic development applicable to all nations has not emerged, some basic factors that limit a poor nation’s economic growth have been suggested. These include insufficient capital formation, a shortage of human resources and entrepreneurial ability, and a lack of infrastructure. Capital Formation Almost all developing nations have a scarcity of capital relative to other resources, especially labor. The small stock of physical capital (factories, machinery, farm equipment, and other productive capital) constrains labor’s productivity and holds back national output. Jeffrey Sachs, a professor at the Earth Institute at Columbia and a key economist in helping to develop the MDG, emphasizes the role of capital in moving countries out of poverty.3 Faced with bad climates, few resources and disease, poor countries find it hard to amass the capital needed to develop. They are stuck in a “poverty trap,” sometimes also called the vicious circle of poverty. Without investment, the capital stock does not grow, the income remains low, and the vicious circle is complete. Poverty becomes self-perpetuating. Sachs argues that one can use foreign aid as a lever to move countries out of poverty, providing the key capital needed for both public and private investments. Indeed, Sachs estimates that $195 billion in foreign aid per year could eliminate global poverty in 20 years. Other economists are less confident that foreign aid can play this role. Both William Easterly, Director of NYU’s Development Research Institute, in his book The Elusive Quest for Growth4 and Dambisa Moyo, a Zambian economist, in her book Dead Aid5 argue that foreign aid can actually hamper development by distorting market incentives for local entrepreneurs. There are also questions surrounding the assumption that poor countries cannot generate capital themselves. Japanese GDP per capita in 1900 was well below that of many of today’s developing nations, yet today it is among the developed nations. Among the many nations with low levels of capital per capita,
|
some—like China—have managed to grow and develop in the last 20 years, whereas others remain behind. In even the poorest countries, there remains some capital surplus that could be harnessed if conditions were right. Many current observers believe that scarcity of capital in some developing countries may have more to do with a lack of incentives for citizens to save and invest productively than with any absolute scarcity of income available for capital accumulation. Many of the rich in developing countries invest their savings in Europe or in the United States instead of in their own country, which may have a riskier political climate. Savings transferred to the United States do not lead to physical capital growth in the developing countries. The term capital flight refers to the fact that both human capital and financial capital (domestic savings) leave developing countries in search of higher expected rates of return elsewhere or returns with less risk. Government policies in the developing nations—including price ceilings, import controls, and even outright appropriation of private property—tend to discourage investment. There has been increased attention to the role that financial institutions, including accounting systems and property-right rules, play in encouraging domestic capital formation. 2Abhijit Banerjee and Esther Duflo, Poor Economics, Perseus Books, 2011. 3Jeffrey Sachs, The End of Poverty: Economic Possibilities for Our Time, Penguin press, NY, 2005 4William Easterly, The Elusive Quest for Growth, MIT Press, 2001. 5Dambisa Moyo, Dead Aid: Why Aid is Not Working and How There Is a Better Way for Africa, Allen Lane 2009. vicious circle of poverty Suggests that poverty is self-perpetuating because poor nations are unable to save and invest enough to accumulate the capital stock that would help them grow. capital flight The tendency for both human capital and financial capital to leave developing countries in search of higher expected rates of return elsewhere with less risk. M20_CASE3826_13_GE_C20.indd 414 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 415 Whatever the causes of capital shortages, it is clear that the absence of productive capital prevents income from rising in any economy. The availability of capital is a necessary, but not a sufficient, condition for economic growth. Other ingredients are required to achieve economic progress. Human Resources and Entrepreneurial Ability Capital is not the only factor of production required to produce output. Labor is equally important. To be productive, the workforce must be healthy. Disease today is
|
the leading threat to development in much of the world. In 2015, about 429,000 people died of malaria, almost all of them in Africa. HIV/AIDS was still responsible for 1 million deaths in 2016, again mostly in Africa, and has left Africa with more than 14 million AIDS orphans. Iron deficiency and parasites sap the strength of many workers in the developing world. Control of malaria and HIV/AIDS were MDG goals for 2015 and considerable progress was made, though considerable work still remains to be done. As we saw in Table 20.1, low-income countries also lag behind high-income countries in literacy rates. To be productive, the workforce must be educated and trained. Basic literacy as well as specialized training, for example, can yield high returns to both the individual worker and the economy. Education has grown to become the largest category of government expenditure in many developing nations, in part because of the belief that human resources are the ultimate determinant of economic advance. Nevertheless, in many developing countries, many children, especially girls, receive only a few years of formal education, though some progress has been made in this area. As technology pushes up the wage premium on skilled workers the impact of low literacy rates on a country’s GDP rises. Just as financial capital seeks the highest and safest return, so does human capital. Thousands of students from developing countries, many of whom were supported by their governments, graduate every year from U.S. colleges and universities. After graduation, these people face a difficult choice: to remain in the United States and earn a high salary or to return home and accept a job at a much lower salary. Many remain in the United States. This brain drain siphons off many of the most talented minds from developing countries. It is interesting to look at what happens to the flow of educated workers as countries develop. Increasingly, students who have come from China and India to study are returning to their home countries eager to use their skills in their newly growing economies. The return flow of this human capital stimulates growth and is a signal that growth is occurring. Indeed, development economists have found evidence that in India, schooling choices made by parents for their children respond quite strongly to changes in employment opportunities.6 The connection between growth and human capital is in fact a two-way street. Even when educated workers leave for the developed world, they may contribute to the growth of their home country. Recently, economists have begun studying remittances, compensation sent back from recent immigrants to their families in less developed countries
|
. Although measurement is difficult, estimates of these remittances are approximately $100 billion per year. Remittances fund housing and education for families left behind, but they also can provide investment capital for small businesses. In 2007, it appeared that remittances from illegal immigrants in the United States to Mexico, which had been growing by 20 percent per year, were beginning to fall with tightening of enforcement of immigration rules. Remittances fell further in 2008–2009 with the recession, but have recovered in subsequent years. In 2016, remittances to Mexico from the United States reached an all-time high of $27 billion, accounting for nearly 3 percent of the country’s income.7 In recent years, we have become increasingly aware of the role of entrepreneurship in economic development. Many of the iconic firms in the 19th century that contributed so strongly to the early industrial growth of the United States—Standard Oil, U.S. Steel, Carnegie Steel—were begun by entrepreneurs starting with little capital. In China, one of the top search engines is Baidu, a firm started in 2000 by two Chinese nationals, Eric Xu and Robin Li, and now traded on NASDAQ, as is AliBaba, a major online retailer. Business writers often compare these 6The classic work in this area was done by Kaivan Munshi and Mark Rosenzweig, “Traditional Institutions Meet the Modern World: Caste, Gender, and Schooling Choice in a Globalizing Economy,” American Economic Review, September 2006, 1225–1252. More recent work includes Emily Oster and Bryce Millett, “Do Call Centers Promote School Enrollment? Evidence from India, Journal of Development Economics, September 2013. 7Juan Jose, Li Ng and Carlos Serrano, “Mexico Yearbook of migration and remittances,” 2017. BBVA Research Report. brain drain The tendency for talented people from developing countries to become educated in a developed country and remain there after graduation. M20_CASE3826_13_GE_C20.indd 415 17/04/19 5:34 PM 416 PART V The World Economy Corruption Corruption is one of the biggest barriers to economic development. Development specialists argue that, fueled by low levels of economic development, poverty and illiteracy are the main culprits behind inherently high corruption levels in some countries. But what exactly does it mean to be corrupt? Transparency International (TI), the Berlin-based non-government organization defines corruption as
|
the abuse of entrusted power for private gain. TI classifies corruption into grand, petty, and political. At one end of the spectrum, grand and political corruption are committed by high-ranking policymakers who use their power to enforce laws and policies that benefit leaders and politicians at the expense of public interest. At the other end, petty corruption is committed by low and mid-level public officials in the process providing public goods and services to ordinary citizens. This could range from receiving bribes to the embezzlement of public funds. Over the last two decades, TI has been gauging corruption using the Corruption Perceptions Index (CPI), which ranks countries based on opinion surveys of the perceptions of experts and businesses. The CPI uses a scale of 0 to 100, where 0 is highly corrupt and 100 is non-corrupt. Most developing nations score below 40.1 The Washington-based organization, Global Financial Integrity (GFI), further includes illegal and underground activities in its definition of corruption. Examples of such activities include drug trading, counterfeit and sub-standard medication, ivory smuggling, people trafficking, theft of oil and natural resources, and piracy. The damage caused by such activities is not only economic, in the form of forfeited tax revenues and capital flight, but also social. This damage has been quantified by the UN Economic Commission for Africa (ECA), which stipulates that the losses due to corrupt activities are profound in Africa. At $50 billion per annum, this loss is nearly double the official development assistance that African nations receive.2 Other spillover effects of corrupt transactions include organized crime and illegal migration across borders. The Organization for Economic Cooperation and Development (OECD) highlights the need for cooperation between Western and African nations to fight corruption. Some of the structural and economic reforms needed are better governance, higher accountability, more jobs, enhanced surveillance, and the organization of informal sectors. CRITICAL THINKING 1. Go to the website of TI, GFI, ECA, or the OECD. What are the recommended policies for developing nations to break the vicious circle between corruption and underdevelopment? 1Transparency International (2018). Corruption Perceptions Index 2017, February 15, Transparency International, Berlin. Retrieved from: https://www. transparency.org/whatwedo/publication/corruption_perceptions_index_2017 2UN Economic Commission for Africa (2018). IFF Background, ECA. https:// www.uneca.org/pages/iff-background Corruption Perceptions Index 2017 Highly Corrupt Very Clean 0
|
–9 10–19 20–29 30–39 40–49 50–59 60–69 70–79 80–89 90–100 Source: Transparency International (2018). Corruption Perceptions Index. M20_CASE3826_13_GE_C20.indd 416 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 417 social overhead capital Basic infrastructure projects such as roads, power generation, and irrigation systems. two firms to Google and Amazon. Providing opportunities and incentives for creative risk takers seems to be an increasing part of what needs to be done to promote development. The work by Easterly and Mayo cited earlier both focus on the potential for poorly focused foreign aid to distort local entrepreneurial incentives and hamper economic growth. Infrastructure Capital Anyone who has spent time in a developing nation knows how difficult it can be to carry on everyday life. Problems with water supplies, poor roads, frequent electrical power outages (in the few areas where electricity is available), and often ineffective mosquito and pest control make life and commerce difficult. In any economy, developing or otherwise, the government plays an investment role. In a developing economy, the government must create a basic infrastructure—roads, power generation, and irrigation systems. Such projects, sometimes referred to as social overhead capital, often cannot be successfully undertaken by the private sector. Many of these projects operate with economies of scale, which means they can be efficient only if they are very large, perhaps too large for any private company or group of companies to carry out. In other cases, the benefits from a development project, although extraordinarily valuable, cannot be easily bought and sold. The availability of clean air and potable water are two examples. Here government must play its role before the private sector can proceed. For example, some observers have recently argued that India’s growth prospects are being limited by its poor rail transport system. Goods from Singapore to India move easily over water in less than a day, but they can take weeks to move from port cities to supply factories in the interior. China, by contrast, spent the bulk of its stimulus money in the 2008–2009 period trying to build new transportation networks in part because the government understood how key this social overhead capital was to economic growth. In 2017, the Chinese, big investors in Africa, finished a 450 mile railway from Addis Abbaba in Ethiopia to Dijbouti, designed to reduce transport costs and fuel economic growth. The Economics in Practice box on page 419 describes one
|
of the unexpected results of government infrastructure provision in Bangladesh. To build infrastructure generally requires public funding. Many less-developed countries struggle with raising tax revenues to support these projects. In the last few years, Greece has struggled to repay its debt partly because of widespread tax evasion by its wealthiest citizens. In many less-developed countries, corruption limits the public funds available for productive government investments, as the Economics in Practice box on page 416 suggests. Strategies for Economic Development MyLab Economics Concept Check Despite many studies, looking across hundreds of countries, there has emerged no consensus on the right strategy to move a country out of poverty. Nevertheless, there are several promising strategies that may prove useful at the country level in some contexts. The Role of Government In the modern capitalist world most investment capital is supplied to entrepreneurs by third parties, either through the banking system we described in previous chapters or through the stock market. For those markets to work, to enable capital to flow, requires trust. Developing this trust in an environment in which most investment is impersonal in turn requires some government oversight. Rules need to be set and enforced, governing the kinds of data reported in financial statements, the way deposits are protected, and terms of loans enforced. The government similarly plays a role in property protections needed in a modern impersonal economy. These institutions are a necessary complement to economic development. The Economics in Practice box on page 419 describes the way in which family loans partially substitute for impersonal loans in Bangladesh where financial institutions are less well developed. Between 1991 and 1997, U.S. firms entered Eastern Europe in search of markets and investment opportunities and immediately became aware of a major obstacle. The institutions that make the market function relatively smoothly in the United States did not exist in Eastern Europe. The banking system, venture capital funds, the stock market, the bond market, commodity exchanges, brokerage houses, investment banks, and so on, have developed in the United States over hundreds of years, and they could not be replicated overnight in the formerly Communist world. Similar problems exist today in the Chinese economy. Although the Chinese equity market has grown rapidly in the last decade, that growth has been accompanied by problems with weak governance and lack of transparency. These issues discourage investments by western firms. M20_CASE3826_13_GE_C20.indd 417 17/04/19 5:34 PM 418 PART V The World Economy Many market-supporting institutions are so basic that Americans take them for granted. The institution of private property, for example, is a
|
set of rights that must be protected by laws that the government must be willing to enforce. Suppose the French hotel chain Novotel decides to build a new hotel in Moscow or Beijing. Novotel must first acquire land. Then it will construct a building based on the expectation of renting rooms to customers. These investments are made with the expectation that the owner has a right to use them and a right to the profits that they produce. For such investments to be undertaken, these rights must be guaranteed by a set of property laws. This is equally true for large business firms and for local entrepreneurs who want to start their own enterprises. China’s ambiguous property rights laws may also be problematic. Although farmers can own their own homes, for example, all rural land is collectively owned by villages. Farmers have the right to manage farmland, but not own it. As a result, transfer of land is difficult. Similarly, the law must provide for the enforcement of contracts. In the United States, a huge body of law determines what happens if you break a formal promise made in good faith. Businesses exist on promises to produce and promises to pay. Without recourse to the law when a contract is breached, contracts will not be entered into, goods will not be manufactured, and services will not be provided. Protection of intellectual property rights is also an important feature of developed market economies. When an artist puts out a record, the artist and his or her studio are entitled to reap revenues from it. When Apple developed the iPod, it too earned the right to collect revenue for its patent ownership. Many less-developed countries lack laws and enforcement mechanisms to protect intellectual property of foreign investments and their own current and future investors. The lack of protection discourages trade and home-grown invention. Another seemingly simple matter that turns out to be quite complex is the establishment of a set of accounting principles. In the United States, the rules of the accounting game are embodied in a set of generally accepted accounting principles (GAAP) that carry the force of law. Companies are required to keep track of their receipts, expenditures, assets, and liabilities so that their performance can be observed and evaluated by shareholders, taxing authorities, and others who have an interest in the company. If you have taken a course in accounting, you know how detailed these rules have become. Imagine trying to do business in a country operating under hundreds of different sets of rules. That is what happened in Russia during its transition. It is clear that economic development requires these financial and legal institutions. There is more
|
debate about how much the lack of these institutions plays a role in keeping some countries poor. Work by Acemoglu, Johnson, and Robinson looking at the history of the African nations assign a prominent role to the lack of institutions in some nations as a cause of poverty.8 Other work suggests that institutions naturally develop alongside of markets and the economy and thus their absence marks market failure rather than causing it.9 The Movement from Agriculture to Industry Consider the data in Table 20.2. The richest countries listed—the United States, Japan, and Korea—generate much of their GDP in services, with little value contributed by agricultural production. The poorest countries, on the other hand, have substantial agricultural sectors, although as you can see, the service sector is also large in a number of these economies. The transition to a developing economy typically involves a movement away from agriculture. Recent work has documented the higher productivity of workers in the nonagricultural sector versus the agricultural sector in developing countries. Even carefully adjusting for difference in human capital of labor in the two sectors, value added per worker is much higher in the nonagricultural sector.10 This tells us that these countries would be better off in terms of productivity if they could more quickly move workers out of the agrarian areas and to the urban work place. Indeed, Gharad Bryan from the London School of Economics and Melanie Morton, of Stanford estimated that almost 20 percent of Indonesia’s growth between 1976 and 2012 could be accounted for by the reductions in migration costs that occurred during the 8Daron Acemoglu, Simon Johnson and James Robinson, “The Colonial Origins of Comparative Development: An Empirical investigation,” American Economic Review, 2001, 1369–1401 9Edward Glaeser, Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer, “Do Institutions Cause Growth?” Journal of Economic Growth, September 2004. 10Douglas Gollin, David Lagakos and Michael Waugh, “The Agricultural Productivity Gap,” Quarterly Journal of Economics, 2014, 939–993 M20_CASE3826_13_GE_C20.indd 418 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 419 Who You Marry May Depend on the Rain In Bangladesh, as in many other low-lying countries, river flooding often leaves large swaths of land under water for substantial portions of the
|
year. By building embankments on the side of the river, governments can extend the growing season, allowing several seasons of crops. The result is a wealth increase for people living in affected rural areas. In a recent paper, several economists traced through some unusual consequences of increasing the wealth of rural populations by creating embankments.1 In Bangladesh, marriages require dowries, paid by the bride’s family to the groom. For poor families, raising these dowries can be difficult and it is not easy to marry now and promise a dowry-by-installment later on. Making people live up to their promises and pay debts is no easier in Bangladesh than it is elsewhere in the world! The result? In hard times and among the poorer families, people in Bangladesh often marry cousins; promises within an extended family are more easily enforced and wealth sharing inside families is also more common. Now let us think about what happens when the government builds a flood embankment, allowing farmers on one side of the embankment to till the land over most of the year, while those on the other side are faced with six-month flooding. Farmers on the flooded side of the river continue to use marriage within the extended family as a strategy to essentially provide dowries on credit. For those farmers on the more stable side of the river, cousin marriages fell quite substantially. Marriage of cousins can have health risks, thus investments in rural infrastructure can have unforeseen positive effects in an area. CRITICAL THINKING 1. What do you think happens to the overall marriage rate as a result of the embankment? 1Ahmed Mushfiq Mobarak, Randall Kuhn, and Christina Peters, “Consanguinity and Other Marriage Market Effects of a Wealth Shock in Bangladesh,” Demography, forthcoming 2013. period.11 Similar results were found in an experiment which gave random subsidies to workers in Bangladesh to outmigrate from the farm area to the city during the lean period of the farm year.12 This work suggests that one way to improve growth in a developing country is to invest TABLE 20.2 The Structure of Production in Selected Developed and Developing Economies, 2008 Per-Capita Gross National Income (GNI) Agriculture Industry Services Percentage of Gross Domestic Product $ 460 570 3,040 3,640 4,640 7,490 21,430 37,840 47,890 30 19 11 12 8 6 3 1 1 23 29 47 44 35 28 36 27 21 47 52 40 44 57 66 61 71
|
78 Country Tanzania Bangladesh China Thailand Colombia Brazil Korea (Rep.) Japan United States Source: The World Bank. 11Gharad Bryan and Melanie Morton, “The Aggregate Productivity effects of Internal Migration: Evidence from Indonesia,” Journal of Political Economy, forthcoming. 12Gharad Bryan, Shymal Chowdhury and Ahmed Mushfiq Mobarak, “Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh,” Econometrica, 2014. M20_CASE3826_13_GE_C20.indd 419 17/04/19 5:34 PM 420 PART V The World Economy import substitution An industrial trade strategy that favors developing local industries that can manufacture goods to replace imports. export promotion A trade policy designed to encourage exports. in transportation networks or other mechanisms to reduce the costs of moving between rural and urban areas. Exports or Import Substitution? Trade strategy is often a central part of a country’s development program. Development economists have historically described two alternatives for the sectors a developing country might wish to support with government policy: import substitution or export promotion. While modern development economists typically take a more inclusive approach to growth initiatives by the government, it is still useful to highlight the differences in these two polar cases. Import substitution is a strategy used to develop local industries that can manufacture goods to replace imports. If fertilizer is imported, import substitution calls for a domestic fertilizer industry to produce replacements for fertilizer imports. This strategy gained prominence throughout South America in the 1950s. At that time, most developing nations exported agricultural and mineral products, goods that faced uncertain and often unstable international markets. Under these conditions, the call for import substitution policies was understandable. Special government actions, including tariff and quota protection and subsidized imports of machinery, were set up to encourage new domestic industries. Multinational corporations were also invited into many countries to begin domestic operations. Most economists believe that import substitution strategies have failed almost everywhere they have been tried. With domestic industries sheltered from international competition by high tariffs (often as high as 200 percent), major economic inefficiencies were created. For example, Peru has a population of approximately 29 million, only a tiny fraction of whom can afford to buy an automobile. Yet at one time, the country had five or six different automobile manufacturers, each of which produced only a few thousand cars per year. The cost per car was much higher than it needed to be because there are substantial economies of scale in automobile production, and valuable resources
|
that could have been devoted to another, more productive, activity were squandered producing cars. As an alternative to import substitution, some nations have pursued strategies of export promotion. Export promotion is the policy of encouraging exports. As an industrial market economy, Japan was a striking example to the developing world of the economic success that exports can provide. Japan had an average annual per-capita real GDP growth rate of roughly 6 percent per year from 1960 to 1990. This achievement was, in part, based on industrial production oriented toward foreign consumers. Several countries in the developing world have attempted to emulate Japan’s early success. Starting around 1970, Hong Kong, Singapore, Korea, and Taiwan began to pursue export promotion of manufactured goods with good results. Other nations, including Brazil, Colombia, and Turkey, have also had some success at pursuing an outward-looking trade policy. China’s growth has been mostly export-driven as well. Government support of export promotion has often taken the form of maintaining an exchange rate favorable enough to permit exports to compete with products manufactured in developed economies. For example, many people believe China has kept the value of the yuan artificially low. Because a “cheap” yuan means inexpensive Chinese goods in the United States, sales of these goods increased dramatically. As Joseph Stiglitz recently pointed out, enthusiasm for export-led manufacturing growth has diminished somewhat as the share of manufacturing in world GDP has diminished. 13 In the case of Africa, a number of economists and policy leaders see promise in improving productivity in agriculture as a path to economic growth. A big issue for countries growing or trying to grow by selling exports on world markets is free trade. African nations in particular have pushed for reductions in tariffs imposed on their agricultural goods by Europe and the United States, arguing that these tariffs substantially reduce Africa’s ability to compete in the world marketplace. Microfinance In the mid-1970s, Muhammad Yunus, a young Bangladeshi economist created the Grameen Bank in Bangladesh. Yunus, who trained at Vanderbilt University and was a former professor at Middle Tennessee State University, used this bank as a vehicle to introduce microfinance to the developing world. In 2006, Yunus received a Nobel Peace Prize for his work. 13Joseph Stiglitz, “From Manufacturing Led Export Growth to a 21st Century Inclusive Growth Strategy for Africa,” Speech in Capetown South Africa, November 15, 2017. M20_CASE3826_13_GE_C20.
|
indd 420 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 421 Microfinance is the practice of lending very small amounts of money, with no collateral, and accepting small savings deposits.14 It is aimed at introducing entrepreneurs in the poorest parts of the developing world to the capital market. By 2002, more than 2,500 institutions were making these small loans, serving more than 60 million people. Two thirds of borrowers were living below the poverty line in their own countries, the poorest of the poor. Yunus, while teaching economics in Bangladesh, began lending his own money to poor households with entrepreneurial ambitions. He found that with even small amounts of money, villagers could start simple businesses: bamboo weaving or hair dressing. Traditional banks found these borrowers unprofitable: The amounts were too small, and it was too expensive to figure out which of the potential borrowers was a good risk. With a borrower having no collateral, information about his or her character was key but was hard for a big bank to discover. Local villagers, however, typically knew a great deal about one another’s characters. This insight formed the basis for Yunus’s microfinance enterprise. Within a village, people who are interested in borrowing money to start businesses are asked to join lending groups of five people. Loans are then made to two of the potential borrowers, later to a second two, and finally to the last. As long as everyone is repaying their loans, the next group receives theirs, but if the first borrowers fail to pay, all members of the group are denied subsequent loans. What does this do? It makes community pressure a substitute for collateral. Moreover, once the peer-lending mechanism is understood, villagers have incentives to join only with other reliable borrowers. The mechanism of peer lending is a way to avoid the problems of imperfect information described in a previous chapter. The Grameen model grew rapidly. By 2002, Grameen was lending to two million members. Thirty countries and 30 U.S. states have microfinance lending copied from the Grameen model. Relative to traditional bank loans, microfinance loans are much smaller, repayment begins quickly, and the vast majority of the loans are made to women (who, in many cases, have been underserved by mainstream banks). A growing set of evidence shows that providing opportunities for poor women has stronger spillovers in terms of improving the welfare of children than does providing comparable opportunities for men. More recently small deposit savings accounts have
|
also been introduced to the under-banked populations in the developing world. Although the field of microfinance has changed considerably since Yunus’s introduction and many people question how big a role it will ultimately play in spurring major development and economic growth, it has changed many people’s views about the possibilities of entrepreneurship and access to financial institutions more generally for the poor of the world. Two Examples of Development: China and India MyLab Economics Concept Check China and India provide two interesting examples of rapidly developing economies. Although low per-capita incomes still mean that both countries are typically labeled developing as opposed to developed countries, many expect that to change in the near future. In the 25-year period from 1978 to 2003, China grew on average, 9 percent per year, a rate faster than any other country in the world. Even during the 2008–2009 U.S. recession, China continued to grow, and it has continued to do so. While India’s surge has been more recent, in the last eight years, it too has seen annual growth rates in the 6 to 8 percent range. Many commentators expect India and China to dominate the world economy in the 21st century. How did these two rather different countries engineer their development? Consider institutions: India is a democratic country, has a history of the rule of law, and has an Englishspeaking heritage—all factors typically thought to provide a development advantage. China is still an authoritarian country politically, and property rights are still not well established—both characteristics that were once thought to hinder growth. Both China and India have embraced free-market economics, with China taking the lead as India has worked to remove some of its historical regulatory apparatus. What about social capital? Both India and China remain densely populated. Although China is the most populous country in the world, India, with a smaller land mass, is the more densely 14An excellent discussion of microfinance is contained in Beatriz Armendariz de Aghion and Jonathan Morduch, The Economics of Microfinance, (MIT Press, 2005.) M20_CASE3826_13_GE_C20.indd 421 17/04/19 5:34 PM 422 PART V The World Economy Boosting Agricultural Income Through Digital Finance Poverty in India is a major issue, especially in rural areas, where around two-thirds of the country’s population lives. Around half of India’s population is employed in the agriculture sector, and more than 58 percent of rural households are
|
dependent on agriculture as their primary source of livelihood. Most of the rural population in India is engaged in smallholder subsistence agricultural production. Mostly consisting of small farmers, this demographic is unlikely to engage with agriculture-related technology or information. Moreover, such farmers do not have adequate access to credit, insurance, or marketing services. If these services were to be provided properly, the levels of crop productivity and profitability are apt to rise. In an effort to provide adequate and easily accessible financial infrastructure, the Indian government is pursuing efforts to digitalize the agriculture sector. A primary reason behind this strategy is that around 70 percent of farmers in India own a mobile phone, and this allows the government to offer digital finance to this sector. Through digital finance, mobile payments provide a secure and cost-effective method for financial transactions in the agricultural sector, particularly for smallholder farms. Mobile phones can connect farmers through a digital platform to help them consolidate their bargaining power and sell their crops at better prices to buyers all over India. Farmers can also secure immediate reimbursement from buyers as they use mobile wallets instead of cash. Similarly, banks dedicated to rural credit and microfinance can benefit from lower administrative costs and fewer cash transactions. This, in turn, can give unbanked farmers access to loans as lenders can start to assess credit risk. Moreover, the Government would be able to increase its resources through tax collection as all payments and transactions can now be traceable. Many developing nations have successfully designed schemes to use mobile phones to generate and disseminate information from geo-imaging services, akin to Google Earth, to warn farmers against weather conditions such as droughts, storms, or heat waves. A 2016 report from the McKinsey Global Institute found that the full adoption of digital finance could increase the collective GDPs of all developing economies by a total of €3 trillion and create up to 95 million jobs by 2025. CRITICAL THINKING 1. With the help of aggregate demand and aggregate supply diagrams, explain the impact of digital finance on the price and quantity of agricultural products. populated. Nevertheless, as is true in most developing nations, birth rates in both countries have fallen. Literacy rates and life expectancy in China are quite high, in part a legacy from an earlier period. India, on the other hand, has a literacy rate that is less than that of China’s and a lower life expectancy. In terms of human capital, China appears to have the edge, at least for now. What about the growth strategies used by the two countries? China has
|
adopted a pragmatic, gradual approach to market development, sharply in contrast to that adopted some years ago in Poland. China’s approach has been called moshi guohe, or “crossing the river by feeling for stepping stones.” In terms of sector, most of China’s growth has been fueled by manufacturing. The focus on manufacturing is one reason that China’s energy consumption and environmental issues have increased so rapidly in the last decade. In India, services have led growth, particularly in the software industry. In sum, it is clear from comparing India and China that there is no single recipe for development. 20.3 LEARNING OBJECTIVE Discuss the intervention methods used by development economists. Development Interventions In the last 20 years, development economists have increasingly turned to much narrower, more microeconomically oriented programs to see if they can figure out which interventions help the condition of the bottom of the income distribution in developing countries. This work has moved away from a search for general recipes for growth and development. M20_CASE3826_13_GE_C20.indd 422 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 423 Random and Natural Experiments: Some New Techniques in Economic Development MyLab Economics Concept Check Suppose we were trying to decide whether it was worthwhile in terms of student achievement to hire another teacher to reduce the student–faculty ratio. One traditional way we might try to answer that question is to find two classrooms with different enrollments in otherwise similar school systems and look at the educational performance of the students. We see comparisons of this sort everyday in newspaper discussions of policies, and many research projects take a variant of this approach, but the approach is subject to serious criticism. It is possible that differences in the two classrooms beyond the enrollment numbers also matter to performance—differences we have failed to correct in the comparisons we make. Crowded classrooms may be in poorer areas (indeed, this may account for the crowding); they may have less effective teachers; they may lack other resources. In the social sciences, it is difficult to ensure that we have comparisons that differ only in the one element in which we are interested. The fact that our interventions involve people makes it even harder. In the case of the classrooms with small enrollment, it may well be that the most attentive parents have pushed to have their children in these classrooms, believing them to be better. Perhaps the best teachers apply to lead these classrooms, and their higher
|
quality makes it more likely that they get their first choice of classrooms. If either of these things happens, the two classrooms will differ in systematic ways that bias the results in favor of finding better performance in the smaller classrooms. More attentive parents may provide home support that results in better test outcomes for their children even if the classrooms are crowded. Better teachers improve performance no matter how crowded the classrooms are. Moreover, in many cases, the differences in the two classrooms—like more attentive parents and teachers—may be impossible to measure. Problems of this sort, sometimes called selection bias, plague social science research. In recent years, a group of development economists began using a technique borrowed from the natural sciences, the random experiment, to try to get around the selection problem in evaluating interventions. Instead of looking at results from classrooms that have made different choices about class size or textbooks, for example, the experimenters randomly assign otherwise identical-looking classes to either follow or not follow an intervention. Students and teachers are not allowed to shift around. By comparing the outcomes of large numbers of randomly selected subjects with control groups, social scientists hope to identify effects of interventions in much the same way natural scientists evaluate the efficacy of various drugs. The leading development group engaged in random experiments in the education and health areas is the Poverty Research Lab at MIT, run by Esther Duflo and Abhijit Banerjee. By working with a range of nongovernmental organizations (NGOs) and government agencies in Africa, Latin America, and Asia, these economists have looked at a wide range of possible investments to help improve outcomes for the poorest of the poor. Of course, not all policies can be evaluated this way. Experimenters do not always have the luxury of random assignment. An alternative technique is to rely on what have been called natural experiments to mimic the controlled experiment. Suppose I am interested in the effect of an increase in wealth on the likelihood that a poor family will enroll its daughters in school. Comparing school behavior of rich and poor families is obviously problematic because they are likely to differ in too many ways to control adequately. It is also not feasible to substantially increase the wealth of a large number of randomly selected parents, but in an agrarian community we may observe random, annual weather occurrences that naturally lead to occasional years of plenty, and by observing behavior in those years versus other years, we may learn a good deal. The weather in this case has created a natural experiment. Empirical development economics thus has added experimental methods to its tool kit as a way
|
to answer some of the difficult and important questions about what does and does not work to improve the lot of the poor in developing nations. We turn now to look at some of the recent work in the fields of education and health, focusing on this experimental work, to provide some sense of the exciting work going on in this field. Education Ideas MyLab Economics Concept Check As we suggested, human capital is an important ingredient in the economic growth of a nation. As economies grow, returns to education also typically grow. As we move from traditional agrarian economies to more diversified and complex economies, the advantages to an random experiment (Sometimes referred to as a randomized experiment) A technique in which outcomes of specific interventions are determined by using the intervention in a randomly selected subset of a sample and then comparing outcomes from the exposed and control groups. natural experiment Selection of a control versus experimental group in testing the outcome of an intervention is made as a result of an exogenous event outside the experiment itself and unrelated to it. M20_CASE3826_13_GE_C20.indd 423 17/04/19 5:34 PM 424 PART V The World Economy individual from education rise. So if we want a nation’s poor to benefit from growth, improving their educational outcomes is key. This leads us to one of the central preoccupations of development economists in the last decade or so: Of the many investments one could make in education, which have the highest payoffs? Is it better to invest in more books or more teachers? How much does the quality of teachers matter? Are investments most important in the first years of education or later? In a world with limited resources in which educational outcomes are very important, getting the right answers to these questions is vital. For most middle-class U.S. students, it may come as a surprise that in the developing world, teacher absenteeism is a serious problem. A recent study led by researchers from the World Bank found, for example, that on an average day, 27 percent of Ugandan and 25 percent of Indian teachers are not at work. Across six poor countries, teacher absences averaged 19 percent. The Poverty Research Lab has conducted a number of experiments in a range of developing countries to see how one might reduce these absences. One successful intervention was introduced in Rajasthan, India, by an NGO called Seva Mandir. Each day when he or she arrived, the teachers in half of Seva Mandir’s 160-single teacher schools
|
were asked to have their picture taken with the children. Cameras were date-stamped. This evidence of attendance fed into the compensation of the teacher. Teacher absentee rates were cut in half relative to the seemingly identical classrooms in which no cameras were introduced. Student absenteeism is also a problem throughout the developing world, reducing educational outcomes even when schools are well staffed with qualified teachers. Several countries, including Mexico, have introduced cash payments to parents for sending their children to school regularly. Since the Mexican government introduced these payments over time, in ways not likely to be related to educational outcomes, researchers could compare student absenteeism across seemingly identical areas with and without the cash incentives as a form of natural experiment. There is some evidence that cash payments do increase school attendance. Natural experiments have also been used to look at the effect of industrialization that improves educational returns as a way to induce better school attendance; the results have been positive. Work using experiments, both natural and random, is still at an early stage in development economics. Although many reform ideas have proven helpful in improving educational outcomes in different developing countries, it has proven hard up to now to find simple answers that work across the globe. Nevertheless, these new techniques appear to offer considerable promise as a way of tackling issues of improving education for the poor of the developing world. Health Improvements MyLab Economics Concept Check Poor health is a second major contributor to individual poverty. In the developing world, estimates are that one quarter of the population is infected with intestinal worms that sap the energy of children and adults alike. Malaria remains a major challenge in Africa, as does HIV/AIDS. In the case of many interventions to improve health, human behavior plays an important role, and here is where development economics has focused. For many diseases, we have workable vaccines, but we need to figure out how to encourage people to walk to health clinics or schools to get those vaccines. We want to know if charging for a vaccine will substantially reduce uptake. For many waterborne diseases, treatment of drinking water with bleach is effective, but the taste is bad and bleach is not free. How do we induce usage? Treated bed nets can reduce malaria, but only if they are properly used. In each of these cases, there are benefits to the individual from seeking treatment or preventive care, but also costs. In the last several years, a number of development economists have explored the way in which individuals in developing economies have responded to policies that try to change these costs and benefits. Intestinal worms, quite common in areas of
|
Africa with inadequate sanitation, are treatable with periodic drugs at a relatively low cost. Michael Kremer and Ted Miguel, working with the World Bank, used random experiments in Kenya to examine the effect of health education and user fees on families’ take-up of treatment of their children. Kremer and Miguel found a number of interesting results very much in keeping with economic principles. First, a program of charging user fees—even relatively low ones—dramatically reduced treatment rates. The World Bank’s attempts to make programs more financially self-sustaining, if used in this area, were likely to have large, adverse public health effects. Elasticities were well above one. Kremer and Miguel also found that as the proportion of vaccinated people in a village grew, and thus the risk of contagion M20_CASE3826_13_GE_C20.indd 424 17/04/19 5:34 PM CHAPTER 20 Economic Growth in Developing Economies 425 fell, fewer people wanted treatment, indicating some sensitivity to costs and benefit calculations by the villagers. Disappointingly, health education did not seem to make much difference. As with the area of education, much remains for development economists to understand in the area of health and human behavior. Development economics continues to be one of the most exciting areas in economics. S U M M A R Y 1. The economic problems facing the Global South, the developing countries, are often quite different from those confronting industrialized nations. of private property, the law and financial reporting to enable the allocation of capital across unrelated individuals. 20.1 LIFE IN THE DEVELOPING NATIONS: POPULATION AND POVERTY p. 412 2. The UN in its Millennium Development Goals identified a number of areas of concern in the developing world: hunger, literacy, child mortality, maternal mortality, diseases like HIV and malaria, gender equality, and environmental quality. In the 15 years of the program from 2000 to 2015, considerable progress was made on each of the UN’s goals, though many challenges remain. 20.2 ECONOMIC DEVELOPMENT: SOURCES AND STRATEGIES p. 413 3. Almost all developing nations have a scarcity of physical capital relative to other resources, especially labor. The poverty trap or vicious-circle-of-poverty hypothesis says that poor countries cannot escape from poverty because they cannot afford to postpone consumption—that is, to save— to make investments. There is debate as to how widespread the poverty
|
trap is and what the right prescription is to solve the problem. 4. Human capital—the stock of education and skills em- bodied in the workforce—plays a vital role in economic development. 5. Developing countries are often burdened by inadequate infrastructure or social overhead capital, ranging from poor public health and sanitation facilities to inadequate roads, telephones, and court systems. Such social overhead capital is often expensive to provide, and many governments are not in a position to undertake many useful projects because they are too costly. 6. Inefficient and corrupt bureaucracies also play a role in retarding economic development in places. 7. Moving to a sophisticated market economy requires government support and regulation of institutions 8. Evidence indicates that in developing nations labor productivity is considerably higher in the industrial urban setting. Some economists suggest easing migration costs as a strategy for growth. 9. Import-substitution policies, a trade strategy that favors developing local industries that can manufacture goods to replace imports, were once common in developing nations. In general, such policies have not succeeded as well as those promoting open, export-oriented economies. 10. Microfinance—lending small amounts to poor borrow- ers using peer lending groups—has become an important new tool in encouraging entrepreneurship in developing countries. 11. China and India have followed quite different paths in re- cent development. 20.3 DEVELOPMENT INTERVENTIONS p. 422 12. Development economists have begun to use randomized experiments as a way to test the usefulness of various interventions. In these experiments, modeled after the natural sciences, individuals or even villages are randomly assigned to receive various interventions and the outcomes they experience are compared with those of control groups. In the areas of education and health, random experiments have been most prevalent. 13. Development economists also rely on natural experiments to learn about the efficacy of various interventions. In a natural experiment, we compare areas with differing conditions that emerge as a consequence of an unrelated outside force. 14. Many of the newer economic studies focus on under- standing how to motivate individuals to take actions that support policy interventions: to use health equipment properly, to attend schools, to receive vaccinations brain drain, p. 415 capital flight, p. 414 export promotion, p. 420 Global South, p. 412 import substitution, p. 420 natural experiment, p. 423 random experiment, p. 423 social overhead capital, p. 417 vicious circle of poverty, p. 414 MyLab Economics Visit www.pearson.com/mylab/economics to complete these
|
exercises online and get instant feedback. Exercises that update with real-time data are marked with. M20_CASE3826_13_GE_C20.indd 425 17/04/19 5:34 PM 426 PART V The World Economy P R O B L E M S All problems are available on MyLab Economics. 20.1 LIFE IN THE DEVELOPING NATIONS: POPULATION AND POVERTY LEARNING OBJECTIVE: Discuss the characteristics of developing nations. 1.1 [Related to the Economics in Practice on p. 413] A paper released by the World Bank in 2014 states that while economic growth is essential for reducing poverty rates, growth by itself is not enough, and efforts to reduce poverty must be complemented with programs that devote more resources to the extreme poor. According to the paper, as extreme poverty declines, growth by itself tends to be less successful at lifting additional people out of poverty because at this point, many still suffering from extreme poverty find it very difficult to improve their lives. Do you agree with this assessment? Why or why not? What fundamental economic concept seems to be at play here? 1.2 The small West African nation of Equatorial Guinea is designated as an upper middle-income country by the World Bank, with a GNI per capita of more than $7,000 when measured in U.S. dollars. Equatorial Guinea also has a poverty rate of more than 76 percent, one of the highest rates in the world. Life expectancy at birth is only 64 years, and the infant mortality rate is almost 7 percent. Do some research on Equatorial Guinea and try to explain the apparent discrepancies listed above for this high-income country. 20.2 ECONOMIC DEVELOPMENT: SOURCES AND STRATEGIES LEARNING OBJECTIVE: Describe the sources of economic development. 2.1 For a developing country to grow, it needs capital. The major source of capital in most countries is domestic saving, but the goal of stimulating domestic saving usually is in conflict with government policies aimed at reducing inequality in the distribution of income. Comment on this trade-off between equity and growth. How would you go about resolving the issue if you were the president of a small, poor country? 2.2 The GDP of any country can be divided into two kinds of goods: capital goods and consumption goods. The proportion of national output devoted to capital goods determines, to some extent, the nation’s growth rate
|
. a. Explain how capital accumulation leads to economic growth. b. Briefly describe how a market economy determines how much investment will be undertaken each period. c. Consumption versus investment is a more painful conflict to resolve for developing countries. Comment on that statement. d. If you were the benevolent dictator of a developing country, what plans would you implement to increase percapita GDP? 2.3 Poor countries are trapped in a vicious circle of poverty. For output to grow, they must accumulate capital. To accumulate capital, they must save (consume less than they produce). They are poor, so they have little or no extra output available for savings—it must all go to feed and clothe the present generation. Thus they are doomed to stay poor forever. Comment on each step in that argument. 2.4 In China, rural property is owned collectively by the village while being managed under long-term contracts by individual farmers. Why might this be a problem in terms of optimal land management, use, and allocation? 2.5 An offshoot of microfinance that has grown significantly over the past several years is an idea known as crowdfunding. With crowdfunding, individuals, businesses, and communities seek monetary support for ideas or projects from other individuals, primarily over the Internet. Three of the largest and most successful crowdfunding Internet sites are GoFundMe, Kickstarter, and Indiegogo, and while the use of the term “crowdfunding” is relatively new and associated with online sites such as these, the concept has been around for many years, with projects such as the pedestal on which the Statue of Liberty resides being constructed using this style of funding. Do some research on crowdfunding and explain whether you believe crowdfunding is a viable alternative to microfinance in poor countries such as Bangladesh. Which source of peer lending, microfinance or crowdfunding, do you believe would be the most successful at reducing the problem of adverse selection? Why? 2.6 [Related to the Economics in Practice on p. 422] Find another example of the use of cell phones as a way to improve market functioning in a developing economy. 2.7 [Related to the Economics in Practice on p. 416] Corruption in a government is often accompanied by inefficiency in the economy. Why should this be true? 2.8 The distribution of income in a capitalist economy is likely to be more unequal than it is in a socialist economy. Why is this so? Is there a tension between the goal of limiting inequality and the goal of motivating risk taking and
|
hard work? Explain your answer in detail. 2.9 Although brain drain is generally associated with develop- ing countries, the recent debt crisis in Greece has generated an exodus of highly educated human capital from this country. In Greece, college education is paid for by the government, and it is estimated that roughly 10 percent of the country’s college-educated workforce have left the country, a majority of which are less than 40 years of age. What implications does this flight of human capital have on growth prospects for the Greek economy? How does the fact that the government pays for college exacerbate MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M20_CASE3826_13_GE_C20.indd 426 17/04/19 5:34 PM this problem? Do some research to find out what has happened to Greek GDP in recent years and what the forecast is for GDP in the near future, and see if this supports your answer. 2.10 [Related to the Economics in the Practice on p. 419] In addition to fewer marriages within extended families, explain what other positive effects are likely to occur in the rural, flood-prone areas of Bangladesh because of increased government spending on infrastructure projects like the building of river embankments and the resulting increase in wealth of the affected rural population. 2.11 Explain how each of the following can limit the economic growth of developing nations. a. A lack of savings and investment b. Unskilled labor c. A lack of social overhead capital 2.12 You have been hired as an economic consultant for the nation of Ishtar. Ishtar is a developing nation that has recently emerged from a 10-year civil war; as a result, it has experienced appreciable political instability. Ishtar has a serious lack of capital formation, and capital flight has been a problem since before the civil war began. As an economic consultant, what policy recommendations would you make for the economic development of Ishtar? 20.3 DEVELOPMENT INTERVENTIONS LEARNING OBJECTIVE: Discuss the intervention methods used by development economists. 3.1 As the text states, investment in human capital is an important ingredient for a nation’s economic growth. The data in the following table shows the net enrollment rates in primary school as a percentage of the relevant group for 10 developing countries in 1999 and 2016. Go to http://
|
CHAPTER 20 Economic Growth in Developing Economies 427 data.worldbank.org and look up per capita GDP for these 10 countries for 1999 and 2016. (Search for GDP per capita [current $US] data.) Calculate the percent changes in per capita GDP from 1999 to 2016 for these 10 countries. Do the changes in per capita GDP seem to correlate with the changes in enrollment rates? What besides increased enrollment may be responsible for the changes in per capita GDP? Net enrollment rate, primary school, percent of relevant group Country 1999 2016 Percent change Burkina Faso Cote d’Ivoire Djibouti The Gambia Ghana Lesotho Mali Niger Senegal Togo 36 57 27 75 62 59 44 26 55 89 76 88 52 76 87 81 62 64 72 87 111 54 93 1 40 37 41 146 31 -2 Source: The World Bank 3.2 The text mentions that in the developing world, teacher absenteeism is a serious problem, averaging 19 percent across six poor countries. An article in the Journal of Economic Perspectives states that absenteeism of health care workers in five of those countries where data was available averages 35 percent, or almost double the rate of teacher absence. Suggest some ways that developing countries might try to successfully reduce the high absentee rates of health care workers, and any possible problems they may encounter in implementing your suggestions. Source: Nazmul Chaudhury, Jeffrey Hammer, Michael Kremer, Karthik Muralidharan, and F. Halsey Rogers, “Missing in Action: Teacher and Health Worker Absence in Developing Countries,” Journal of Economic Perspectives, 20, no. 1, Winter 2006, pp 91–116 QUESTION 1 Property Rights and the Rule of Law are considered essential for promoting economic growth in developing countries. How do these two features promote economic growth? QUESTION 2 A commonly discussed problem among the global poor is the lack of a credit history. How would this hold poorer households back in terms of increasing their income and wealth? MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M20_CASE3826_13_GE_C20.indd 427 17/04/19 5:34 PM PA RT VI METHODOLOGY 21 Critical Thinking about Research CHAPTER OUTLINE AND LEARNING OBJECTIVES 21.1 Selection Bias p. 429 Give some
|
examples of studies that might suffer from selection bias. 21.2 Causality p. 430 Understand the difference between correlation and causation. 21.3 Statistical Significance p. 437 Understand how researchers decide whether their results are meaningful. 21.4 Regression Analysis p. 438 Understand how regression analysis can be used for both estimation and testing. 428428 Throughout this book we have highlighted the many areas in which economists use data and statistical methods to answer questions that are important to households, businesses, and government policymakers. Some of these questions are narrow: What happens to the sales of ketchup when the manufacturer raises its price? How much will charging a small fee for a vaccine in a developing country affect vaccination rates? Others are much broader: Will a large unexpected fall in housing prices have a substantial effect on household consumption? What happens to employment if we raise the minimum wage? These are all questions that we can begin to answer with economic theory, as you have seen in this text. To get quantitative answers to questions like these, we need to use statistical methods to look at real world data. In this chapter we provide an introduction to the tools economists and other social scientists use to look at data. We will focus both on the standard techniques used and on some of the most common pitfalls associated with using data to answer complex questions. The statistical tools that economists use to analyze issues are an important part of the discipline. If you go on in economics, you will learn much more about these tools. For those of you who do not continue to study economics, we hope the introduction here will allow you to be a more discriminating consumer of the economic research that you see described in the media and elsewhere. The techniques you will learn in this chapter are used in many fields other than economics. Psychology, political science, some historical research, some sports research, and some medical research also use these techniques. M21_CASE3826_13_GE_C21.indd 428 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 429 21.1 LEARNING OBJECTIVE Give some examples of studies that might suffer from selection bias. selection bias Selection bias occurs when the sample used is not random. survivor bias Survivor bias exists when a sample includes only observations that have remained in the sample over time making that sample unrepresentative of the broader population. Selection Bias We all know that people slow down physically as they age. World records in track and field are not set by 45-year-olds.
|
Few professional baseball players continue to play after the age of 45. Yet consider this: In the 2013 Chicago marathon, the average time of the 30- to 39-age group for men was 4 hours and 17 minutes, which was essentially the same as the average time of the 40- to 49age group for men of 4 hours and 18 minutes. What do we make of this? Should we conclude, for example, that in the marathon there is essentially no slowing down in the 10-year age interval between the mid-30s and the mid-40s? Or say you came across a study that randomly sampled 1,000 70-year-old men and 1,000 90-year-old men and measured their bone density. Can we compare the average bone density of the 70-year-olds to that of the 90-year-olds to estimate how much bone density on average declines with age? The answer to both questions is no. In both cases, there is a substantial likelihood of selection bias, which results in unreliable answers. There are many aged 30–39 ham-and-eggers running the Chicago marathon, but many fewer casual runners aged 40–49. Many people aged 30–39 run for fun, to impress a friend, or to pay off a bet. Many of these casual runners have probably selected out by age 40. Moreover, one reason people select out or stop running is that they discover they are not good runners. As a result, the average runner left in the age 40- to 49-interval is likely to be a better runner than the average runner in the age 30- to 39-interval. It is thus not surprising that there is little change in the average times between runners in the two age intervals, but this says nothing about how fast a particular runner slows down with age. We are in some sense comparing apples and oranges in looking at the two groups. Selection bias would also exist in the bone density study. There are fewer 90-year-old men than there are 70-year-olds. Those with lower bone density at age 70 are more likely to have fallen, broken a hip and passed away. The men left in the population age by 90 disproportionately will thus consist of those who at younger ages had higher bone densities. So it would not be sensible to compare the average bone density of the two samples. This comparison would tell us nothing about how bone density changes with age for a particular person. The type of selection bias in these two examples is also called survivor
|
bias, for obvious reasons. The more fit in the populations have survived, so there is a bias in comparing younger and older groups. Similar problems arise in financial markets when we make inferences about corporate returns in the general market from a population of firms that has survived in the market for a long period. Firms that survive are typically different, generally more successful, than the average firm. Apple, which has survived for a number of years, is surely different in its ability to deliver innovative products that people want than the average company. The problem of selection bias pervades many studies in economics (and other disciplines). In recent years there has been considerable interest in trying to understand and improve educational outcomes in the United States. In many areas, charter schools have grown up in part to experiment with alternative educational methods. Charter schools are publicly funded, providing free education to their students, but operate independently of the traditional school district and thus have more autonomy in terms of choices around teacher selection, school hours, and pedagogy. Naturally, there has been considerable interest in how these different charter schools are performing. It might occur to you that one way to answer this question is to compare the scores of students in charter versus traditional schools in an area on the common mastery tests now given across all schools in the United States. Indeed, we see comparisons of this sort often in local newspapers, but here too in making this comparison, you would be running into the problem of selection bias. Where does the bias come in here? In most charter systems, students are randomly chosen to attend the school. You might think this would eliminate the selection bias problem. Unfortunately, that random choice does not fully eliminate the problem. In most charter systems, to be chosen in the lottery you must apply in the first place. Families who apply to a lottery for a charter school may well differ considerably from those who do not apply. Those differences—more attention to education, more organizational skills, and so on—are both likely to matter to educational performance and will be hard for us to observe. In other words, children who apply to a charter school may do better on the mastery tests than the average child, even if he or she does not get chosen to attend the charter! As we will see in a later section, there are ways around this selection problem but they require some ingenuity. M21_CASE3826_13_GE_C21.indd 429 17/04/19 4:29 AM 430 PART VI Methodology One more example may help to show the
|
range of the issues involved. Many studies in the medical area are aimed at helping us figure out how to live longer and healthier lives. Suppose you were interested in the effect on longevity of exercise. Luckily, you found a long-term study that tracked how often people exercised over many years and found that those people who exercised more also lived longer. Should you conclude that exercise in fact increases life span? The answer is again no. In this example we are comparing a group of people who chose to exercise with a group who chose not to. The fact that a group of people do or do not choose to exercise tells us that they likely differ on many other grounds that might independently affect life span. People who choose to exercise also likely make other healthy choices, most of which will be hard for a researcher to observe. So the longevity edge might come from the fact that one group exercised, whereas the other did not, but it might equally have arisen from the fact that the first group consists of people who make healthy choices while the second does not. A common problem in many of these cases is that we are comparing groups who not only engaged in different activities, activities whose effect we seek to measure, but people who made different choices. To the extent that those choices reflect group differences that themselves matter to the outcomes we are measuring, we bias (or distort) our results. In the last few years, economists have become increasingly sensitive to the problem of selection bias and have engaged in many creative ways to try to eliminate the bias problem. We will describe some of the solutions to the bias problem later in this chapter. For now, we hope you will look at some of those newspaper headlines with a more skeptical eye! 21.2 LEARNING OBJECTIVE Understand the difference between correlation and causation. Causality As we have seen, selection bias makes it difficult to identify the effect of a treatment on a population. In other words, selection makes it hard to pin down causal effects. Identifying causality is a general issue in data analysis and goes beyond problems that arise because of selection bias. We will consider a number of causality issues in this section. correlated Two variables are correlated if their values tend to move together. Correlation versus Causation MyLab Economics Concept Check Most people who have blue eyes also have light colored hair. Most people who have minivans also have children. Evidence suggests that people who are obese have a disproportionate number of obese friends. What can we conclude from these facts? Do blue eyes cause blond hair? Do min
|
ivans cause people to have children? Is obesity contagious, caught from one’s friends? When two variables tend to move together, we say they are correlated. If the two variables tend to move in the same direction, we say they are positively correlated, and if they tend to move in opposite directions, we say they are negatively correlated. In the examples above, the variables in each of the three sets are positively correlated; but correlation does not imply causation. It does not take a degree in biology to know that blue eyes do not cause blond hair. Likely evolution has selected simultaneously on these two features causing them to appear together. Dumping a bottle of peroxide on my head, although it will surely make me a blond, will not change my eye color at all! In economics, as well as in other fields, theory is often quite helpful in helping us to differentiate between correlation and causality. Minivans and children provide another example in which we need to sort out correlation and causation. In the data we see that the majority of minivan owners have children. Clearly minivans do not often cause children to be born (“Might as well have a fourth child. We already own a minivan!”). Here it is likely the causality runs in the opposite direction. Minivans are most attractive to families with children. So having children may indeed cause people to buy a minivan. Think now about why getting the causality right matters. If Japan, for example, wants to increase its very low birth rate, giving everyone a free minivan is not likely to be effective. Minivans do not by and large cause people to want children. Knowing the relationship between minivans and children is clearly relevant to automobile manufacturers who will want to exploit this relationship by focusing their marketing campaigns on families. An increase in average family size causes a shift in the demand for large minivans but the reverse is not the case. M21_CASE3826_13_GE_C21.indd 430 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 431 The most complicated of the examples is obesity. Here there are theoretical arguments that support a hypothesis of causality running in both directions. Eating and exercise are social for many people, so having obese (or conversely thin) friends may well have an effect on your own weight. But in some circles at least obesity is a social stigma, and it may well be that being obese limits one�
|
�s choice of friends. Thus, it is plausible that having obese friends does increase your own chances of being obese, but it is also likely that being obese increases your chances of having obese friends. Identifying causality is critical for much policy work. Knowing that early exposure to reading is correlated with high adult incomes is interesting. Knowing that it causes high incomes suggests a policy intervention. Much empirical work in economics is concerned with trying to determine causality, given how important it is for policy issues. Let us consider a few ways researchers have used to identify causation. Random Experiments MyLab Economics Concept Check The gold standard for empirical work is the random experiment that many of you will be familiar with from medical research. If a research team is trying to decide if a particular drug helps in treating some form of cancer, for example, a standard protocol is to randomly divide the patients afflicted with the disease into two groups, provide one group with the drug, and give the other a placebo. With a large enough group, and enough time, one should be able to tell if the drug is effective. (Of course, there is much non-human pretesting for safety reasons). Notice in this protocol that we did not select our samples by asking people to choose whether they wanted to take the drug or not (all agreed to the drug). Indeed, part of a standard medical protocol is that patients do not know which group they are in during the experiment. In this type of experiment, we have no selection issue, since there has been no user selection. Experiments of this sort are also run in economics and are relatively prevalent in the area of economic development. To give an example, suppose we are interested in the effects of class size on educational achievement, say test scores. Comparing classes with large and small enrollments will clearly not be informative. Among other things, it is well known that classes in more affluent areas have smaller classes than those in poorer areas and that affluence will bring with it many advantages that likely lift test scores. Instead, one could run an experiment that randomly assigns students to classes that differ in enrollments and then later compare test scores for the different groups. If the assignments are truly random, there are no selection issues. Although random experiments are common, especially in the medical field, they are not always possible to carry out. Suppose we are interested in the link between smoking and cancer. We can, of course, take a large group of mice, randomly divide them into two groups, expose one but not the other to smoke,
|
and see whether the two groups differ in their cancer incidence. As long as our sample is reasonably large, we should be able to see a difference in cancer rates if a causal relationship between smoking and cancer exists. We have done a random experiment just as we described. Notice how this experiment differs from just comparing cancer rates of smokers to non-smokers. People who smoke have chosen to smoke and may well have made a number of other choices that could be unhealthy. As hard as we try to control for those smoker/ nonsmoker differences, our ability to do so is limited. For the mice there are no choice problems to worry about. If we find that smoking causes cancer in mice; it remains to determine whether the same holds for people. Clearly, we cannot force a randomly chosen group of people to smoke and then see if their cancer rates differ from that of a control group. For many of the questions economists are interested in, it is difficult to use random experiments. Randomly exposing groups of people to something that is potentially harmful is unethical and would not pass a human subjects protocol review. Even if we are looking at interventions that have only potential benefits and no costs, we still face the problem that the randomly chosen subjects we start out with may decide not to join the study or to leave the experiment early. If this happens, the groups left will no longer be random. When there is some discretion among subjects to either take up a treatment offer or to continue in a treatment over time, selection bias will again potentially creep in to our experiment. What do we do under these circumstances? Consider a university that has admitted 200 at-risk students from households with low incomes. It has a summer program before college begins to better help prepare such students for M21_CASE3826_13_GE_C21.indd 431 17/04/19 4:29 AM 432 PART VI Methodology intention to treat A method in which we compare two groups based on whether they were part of an initially specified random sample subjected to an experimental protocol. college life. The university wants to know if this program improves a student’s four-year college performance, say measured by a student’s four-year GPA. How might it proceed? Assume that the university randomly samples 100 of the 200 at-risk students and invites them to attend the summer program at no cost. Say 60 accept the offer and take the summer program. After four years the GPAs of all the 200 students are collected and we learn that the average
|
GPA of the 60 students who took the program was higher than the average GPA of the 140 students who did not take the course. Could you conclude from this that the program had a positive effect? No. Once again, we have a selection issue. While the 100 students offered the program were indeed a random sample, the 60 students who took up the offer were not. Maybe the 60 were on average less talented than the 40 who refused the offer and felt the need to take the program, whereas the more talented 40 did not. Or maybe the 60 were on average more serious students or more organized. However the bias runs, we cannot assume that the 60 students who accepted are a random sample of the 200 initial students. Here we are not even sure if those who accept are better or worse than the non-accepters. That is, we do not know the direction of the bias. The group of the 100 students initially drawn and invited to join the program was random by design. So after four years we can compare the average GPA of the 100 students who were offered the program to the average GPA of the 100 students who were not. If the program has a positive effect, the first average GPA should be greater than the second. You might think this is an odd process for testing the efficacy of the summer program. After all, 40 of the students whose scores we are looking at did not take the program! If they did not take the program, why are they included in the average GPA along with actual program-takers? We include all students who were made the program offer in our test sample to avoid selection bias. This procedure, which is also used in medical experiments that have patient drop-outs, is called intention to treat, but proceeding this way does have a cost. Suppose only 10 students of the 100 took up our offer. In this case, we are comparing two random groups of students, one of which has no one taking the program and the other with 10 in the program and 90 not. With so many non-takers, it will be hard to find any gains from the program. If instead all 100 students invited to the program actually enrolled, clearly we would have more confidence that we could find an effect from the program if any existed. Whatever the case, we need to compare the performance of the 100 offered students with that of the 100 non-offered to avoid selection bias. Notice that intention to treat makes it harder to find results from a treatment and in this sense is a conservative statistical technique. The
|
Economics in Practice box describes an experiment run by the U.S Department of Housing and Urban Development (HUD) using randomly assigned housing vouchers to examine the effects of community on household well being. Here the method of intention to treat is used. Regression Discontinuity MyLab Economics Concept Check In many situations economists do not answer their empirical questions with random experiments, but rather try to make inferences from market data, data that come out of the everyday transactions and choices individuals make. Using market data has a number of advantages: These data reflect real choices made in everyday life by households. Much of the data are collected as a matter of course by either government or business and so are easily available to researchers, but the fact that the data reflect individual choices, done in relatively uncontrolled settings, makes the identification of causality especially difficult. Carefully designed experiments, on the other hand, are expensive. There are a number of procedures that researchers have used to try to make progress in this area. The United States has more prisoners per capita than any other OECD country,1 with roughly two million incarcerated. Many of those released from prison are re-arrested within a short period of time. How does what happens while someone is in prison affect the likelihood they will be re-arrested? Do the conditions in prison affect recidivism rates?2 Arguments about 1The OECD is the Organisation for Economic Co-operation and Development. It consists largely of developed world countries, heavily weighted toward Europe. 2This discussion is based on M. Keith Chen and Jesse Shapiro, “Do Harsher Prison Conditions reduce Recidivism? A Discontinuity-based Approach.” American Law and Economics Review June 2007. M21_CASE3826_13_GE_C21.indd 432 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 433 Moving to Opportunity It is well known that children who grow up in high- poverty areas on average end up as adults with lower educational attainments, poorer health, lower income levels, and a higher likelihood of being incarcerated at some point in their lives. To what extent are these results attributable to the neighborhoods in which these children grow up, and, relatedly, how much could they be changed by a locational change? These are the very central policy questions posed by an experiment run by the U.S. Department of Housing and Urban Development in the mid-1990s and recently re-evaluated by a group of economists.1
|
The Moving to Opportunity program offered to randomly selected families living in high-poverty housing projects housing vouchers that they could use to move to lowerpoverty neighborhoods. The random granting of the vouchers was a direct attempt to avoid the selection bias problems found in earlier studies of housing and later outcomes. It is easy to see that if we simply look at life outcomes for children whose families move out of high-poverty areas to those who remain in those areas we will have serious selection bias issues. Moving families likely have more access to resources—perhaps ones we cannot observe—and perhaps more initiative or organizational ability than those who stay. Those differences might well have an effect on their children’s outcomes independent of the gains from the move. By randomizing the voucher choice, HUD attempted to remove the choice element. Not all families offered the vouchers moved, so the researchers used the intention-to-treat methodology described in the text to control for the potential selection bias. Early results from the experiment found little results on the economic well-being of moving families, though there were gains in mental and physical health. A longer-term, recently completed study by some of the same authors, which looked at tax data, found substantial effects on income levels of those children who were younger than 13 years of age when their families moved, with average gains of 31 percent higher incomes for the young movers. CRITICAL THINKING 1. Some of the same researchers whose work is described also did another study looking at the outcomes of households that moved versus those that did not in the general population. To control for selection bias, the researchers compared children of different ages within families to see how much more time in the better neighborhood influenced younger versus older children. How does this attenuate the selection bias issue? 1Raj Chetty, Nathaniel Hendren, Lawrence Katz, “The Effects of Exposure to Better Neighborhoods on Children: New Evidence for the Moving to Opportunity Experiment,” American Economic Review April 2016, 855–902. this question have been made on both sides. Some argue that harsh conditions reduce recidivism because the worse the conditions, the more incentivized released prisoners will be to stay out of prison. On the other side, harsh prison conditions may increase a taste for violence or reduce a prisoner’s future labor market value. This would suggest that harsh conditions increase recidivism. On questions like this, it is important to bring data and evidence to the table. What happens if we just compare recidivism rates in
|
prisoners from more or less harsh prisons? Here again, identifying causality is problematic. In general, harsher prisons house more serious criminals. So, if we see more recidivism from those coming out of harsher prisons, it could well be that the recidivist traits caused the prison choice, rather than the prison type causing the recidivist traits. Chen and Shapiro used an interesting strategy, called regression discontinuity, to sort out causality. The design works as follows. Once an inmate is convicted and enters the federal prison system, he or she is given a security score. The score predicts the prisoner misconduct and security needs. regression discontinuity Regression discontinuity identifies the causal effects of a policy or factor by looking at two samples that lie on either side of a threshold or cutoff. M21_CASE3826_13_GE_C21.indd 433 17/04/19 4:29 AM 434 PART VI Methodology Control Groups and Experimental Economics Most economic interventions are made in response to an unsatisfactory situation based on inferences from available data. The aim of experimental evaluations, or randomized controlled trials (RCTs), is to establish an ideal comparison group by design from the beginning of an intervention. Government programs often use regression discontinuity to set a threshold for receiving treatment, allowing a researcher to use individuals very close to the threshold as a control group. A 2007 study by economists Esther Duflo, Rachel Glennerster, and Michael Kremer shows that impact assessment is complex and requires a comparison of the current situation of the beneficiaries of a policy to the situation they might have faced in lieu of the policy.1 Participants are randomly assigned to the treatment group or the comparison group, ensuring that there is almost no difference between the two groups. As established in the study, this implies that the outcome is not due to a systematic difference between the two groups that would have existed even without the application of the treatment. Since it is difficult to effectively compare beneficiaries to non-beneficiaries, it is not always possible to derive a useful conclusion of a measure. In fact, it may sometimes have detrimental consequences. According to the study (Duflo et al.), the way a control group is built may have a strong impact on the outcomes. So a decision maker may choose the results that best suit their situation, but derive limited effects. In 2009, Behaghel, Crépon, and Gurgand developed a new test for the effectiveness of a personalized support measure for the unemployed in France.2 The study
|
found that comparable control groups are formed by drawing lots from a qualified population. By establishing two groups, the beneficiary and the non-beneficiary, the researchers could establish the efficiency of a policy and measure its impact on the declining unemployment levels In his article “Short-Run Subsidies and Long-Run Adoption of New Health Products: Evidence from a Field Experiment,” French economist Pascaline Dupas used an experimental approach to test the price elasticity of goods in the health sector in Africa.3 To do so, the difference between free and paid distribution of health products, which is a major debate for health policies, was tested. The approach used the regression discontinuity procedure, and provided 644 households with different vouchers, ranging from 0 to 250 Kenyan shillings, to purchase mosquito nets. The results showed that 98 percent of the population with a voucher for a free mosquito net did get the net while 50 percent of those who received a 50 shillings voucher actually purchased the net. Among those who would have to pay 190 and 250 Kenyan shillings, only 11 percent bought a mosquito net. The conclusion from such a study is that even in the health sector the price elasticity is very high. CRITICAL THINKING 1. Can you think of another situation for which a regression discontinuity technique might be useful? 1Esther Duflo, Rachel Glennerster, and Michael Kremer, “Méthodes d’évaluation des Politiques Publiques [Methods of Evaluation of Public Policies],” The Ministry of Labor, Employment and Health, France, June 2011; and “Using Randomization in Development Economics Research,” Centre for Economic Policy Research, January 2007. 2 Luc Behaghel, Bruno Crépon, J. Guitard, and Marc Gurgand, “Evaluation d’impact de l’accompagnement des demandeurs d’emploi par les op´erateurs priv´es de placement et le programme Cap vers l’entreprise,” [Impact of counselling unemployed jobseekers], September 2009. 3 Pascaline Dupas, “Short-Run and Long-Term Subsidies Adoption of New Health Products: Experimental Evidence from Kenya,” Stanford University, July 2013. There is no personal judgment in creating this score, which simply adds up points depending on the prisoner’s record. The score then determines prison facility based on availability
|
of beds. Scores above six typically go to higher-security (and typically harsher) facilities. Placement also depends on bed space, and this means that prisoners with similar scores may end up in different types of prisons. Regression discontinuity effectively compares outcomes from individuals who are close to either side of a dividing line. In this example, we are effectively comparing recidivism rates for prisoners sent to harsh versus less harsh prisons who were virtually identical on their pre-prison scores. The study in fact found that harsh prisons do not reduce recidivism, but may in fact increase it. Similar methods have been used in other instances in which the existence of a black-and-white line based on a continuous score for individuals determines whether or not an individual is “treated.” Most government programs for unemployment or insurance disability benefits have this property of setting an absolute threshold for receiving treatment, allowing a researcher to essentially use individuals very close to the threshold as a kind of control group. M21_CASE3826_13_GE_C21.indd 434 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 435 difference-in-differences Difference-in-differences is a method for identifying causality by looking at the way in which the average change over time in the outcome variable is compared to the average change in a control group. Difference-in-Differences MyLab Economics Concept Check Another interesting procedure to try to get a better handle on causality in social science studies is called the method of difference-in-differences. Suppose we have a community in which a small nonprofit has run a community gardening program. The group is convinced that this program increases housing values. Someone in the group suggests that they just look at what has happened to housing values in the community in the four years since the program began as a measure of the program’s success. It is easy to see that this will not work. Housing prices are quite volatile, moving with the overall level of economic activity in an area. In other words, much of the fluctuations in housing prices have nothing to do with community gardens. Another suggestion might be to compare the housing prices in this community with those in a similar neighboring community without the program, but this procedure too is problematic, as no two communities are exactly alike. The difference-in-differences method takes a third approach that melds these two ideas. In particular, we try to relate the difference in our community’s housing values
|
over time to the difference in a neighboring community’s values over the same time (hence the name differencein-differences). If all of the other factors that affect housing values are the same between the two communities (that’s why we have chosen a neighboring community), then this difference-indifferences procedure will show us the effect of the program. To be clear on what the procedure does, let pbega and pbegb denote the average housing values in communities a and b before the garden project began in community a. Let penda and pendb denote the average housing values after four years in the two communities. Then the effect of the garden project on housing values in community a is estimated as: effect = penda - pbega - (pendb - pbegb) We take the difference in values in community a and subtract from it the difference in values in community b. The difference-in-differences methodology is reasonably common in the social sciences. There are pitfalls as well in doing this work, pitfalls that come in part from the difficulties of identifying an appropriate comparison group. Consider the following example: Stimulated in part by what has been happening to the cognitive functioning of aging professional athletes, especially in football, there has been growing concern among university leaders about the long-term effects of injuries in college sports. Short of banning football, which some would advocate, there have been other suggestions to reduce the incidence of injury, notably requiring better helmets and/or eliminating kickoffs (with the ball always starting on the 20 yard line). Suppose that several years ago the Ivy League introduced such regulations and that a researcher was interested in seeing whether the regulations in fact had reduced injuries. To test whether the regulations helped, we could compare the average number of injuries per game measured in the year before the new rules, denoted ybeg, with injuries in the year after the new rules were instituted, denoted yend. But as in the case of housing values, we cannot be sure that nothing happened in the world of Ivy League football other than the rule changes over this period. Maybe the NCAA introduced other rule changes for all the colleges in the country, including the Ivy League colleges, which were designed to lessen injuries, such as telling referees to be stricter. We need a comparison group, a second set of differences. One possible comparison group might be the PAC-12 conference. Assume that this conference did not introduce the new rules on helmets and kickoffs. Again, we collect data on
|
the average number of injuries per game for the same two years we used in the Ivy League case for this conference, denoted zbeg and zend. We can then compare the difference between these two values and the difference between the two Ivy League values (difference-in-differences): effect = yend - ybeg - (zend - zbeg) By subtracting the PAC-12 difference from the Ivy difference we are controlling for country-wide changes that occurred during the two years. The variable effect is then the amount attributable to the Ivy League regulations only. M21_CASE3826_13_GE_C21.indd 435 17/04/19 4:29 AM 436 PART VI Methodology Using Difference-in-Differences to Study the Efficacy of Medical Insurance in Japan The hefty pension payments and healthcare expenditure that benefit Japan’s retiree aging population has started to raise red flags among the Japanese. Japan is burdened with swelling government debt that, in 2017, approached 250 percent of GDP. The working population is increasingly pressurizing the Health, Labor and Welfare Ministry to increase social welfare services and income for the working productive generation that is suffering from declining household income due to the financial meltdown and the ensuing wage restraints. Human development experts also argue that the nation should spend more on healthcare and education of the younger and the working generations who are the future of Japan. Among the core elements of income redistribution and human development expenses was the government reducing the citizens’ co-payment share to 10–30 percent of total medical costs. A daunting question facing the government is whether healthcare for the youth has actually had a positive impact on their health. A recent study used a differenceindifferences approach in order to investigate the impact of an increase in the cost sharing rate on medical service utilization among young school pupils with chronic medical conditions.¹ To receive valid results, the study compares 2,896 students, who are divided into two groups. The first group comprises second and third grade beneficiaries of the government- sponsored National Health Insurance. The second consists of the counter factual group of fourth grade children who were not allowed to receive healthcare benefits. The study was conducted for 24 consecutive months from 2012 to 2014. There was no major change in health conditions for both groups. The results reveal that the group that benefited from the cost-sharing rate increased its inpatient service utilization and reduced its outpatient service. Mild and chronic conditions for the other group showed a sharp drop in outpatient service. The study
|
of the behavioral trends will aid policymakers and medical practitioners. But as helpful as the differenceindifference statistical approach is, it is imperative that more specialized medical studies are conducted before deciding on public expenditure on medical services. CRITICAL THINKING 1. How can a non-governmental organization (NGO) that offers a cash support program to the poor use the difference-in-difference approach to assess the effectiveness of its poverty reduction program? 1A. Miyawaki, H. Noguchi, and Y. Kobayashi, 2017. “Impact of Medical Subsidy Disqualification on Children’s Healthcare Utilization: A Difference-in- Differences Analysis from Japan,” Social Science and Medicine, 191, 89–98. This looks neat, but there are several potential pitfalls to this research plan. Most fundamentally, we have assumed that the two-year changes absent the Ivy League regulations are the same for both conferences, but PAC-12 football is not exactly like Ivy League football (ask any serious college sports fan!). Those differences may be important not only in starting levels (which is fine) but in changes over time (which is not fine). PAC-12 football is played at a higher level than the Ivy League, so it could be that the change in its injuries per game over the two years is not a good approximation of what the Ivy League change would have been absent the regulations. Perhaps the PAC-12 coaches pushed their players even harder and this led to increased injuries. If the PAC-12 is not a good comparison group, then difference-in-differences will not work in this case. One more point to reflect on in this football example. With safer helmets it could be that the players play rougher knowing that they are better protected, and playing rougher, other things being equal, increases injuries. Regulations have the potential to affect behavior in ways not anticipated by the regulators. Some of the original work documenting this effect was done by Sam Peltzman, a Chicago economist, who found that seat-belt laws might perversely encourage people to drive faster than they did without seat belts because they felt safer.3 In the helmet case, some gains from the physical protection of a helmet might be offset by the behavioral changes it induces in the intensity of play. Economic research is not an easy task, but we hope you can see that it encourages care and creativity! 3Sam Peltzman, “The effects of automobile safety regulation,” Journal of Political
|
Economy, August 1975. More recent work has questioned this result. M21_CASE3826_13_GE_C21.indd 436 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 437 21.3 LEARNING OBJECTIVE Understand how researchers decide whether their results are meaningful. p-value The probability of obtaining the result that you find in the sample data if the null hypothesis of no relationship is true. statistical significance A result is said to be statistically significant if the computed p-value is less than some presubscribed number, usually 0.05. Statistical Significance We all know that in tossing a coin there is a 50 percent chance we will get heads. Nevertheless, it is not true that coin tosses always alternate between heads and tails. Sometimes we get two or three heads in a row before a tail shows up. How many heads would we need to get in a row before we started to think that there was something wrong with the coin? In the coin example we answer this question by thinking about how likely it is that a fair (or normal) coin would give us heads after heads. Two heads in a row is relatively common, happening 25 percent of the time (0.5 times 0.5). Even four in a row sometimes happens (about 6 percent of the time). However, six heads in a row happens only about one in a hundred times. At that point you may be suspicious about the coin tosser and begin to think this is not a fair coin! In thinking about our results in empirical work in economics we use the same basic logic as we try to figure out what we can conclude from the data we have gathered and the statistical tests we have employed. The key question for the researcher is to figure out if the results he or she has found have occurred “by chance” or if they really mean something. To make that judgement researchers turn to the concept of statistical significance. Return to the example of the summer program experiment and suppose the GPA difference observed after the program was 0.3 on a 4.0 scale. Can we conclude that the program really had a positive effect on GPA, or is 0.3 so small that it was likely due to chance? A common way of looking at this problem is to begin by assuming that the effect of whatever we are testing, here the summer program, is zero and then ask what is the probability we got the result we did if the true effect is zero. The assumption of no effect
|
is called the null hypothesis. In our earlier example, our null hypothesis was that the coin was fair. Here the null hypothesis is that the summer program has no effect on GPA. What is the probability we got a difference in GPA of 0.3 if the null hypothesis is true? The probability that one got the result that one did if the null hypothesis is true can be computed given certain statistical assumptions. It is called a p-value. A small p-value means that the probability is small of getting the result if the null hypothesis is true. If for the 0.3 GPA difference, the p-value was 0.02, this says that there is only a 2 percent chance of getting this value if the summer program truly has no effect on GPA. The term statistical significance is commonly applied to a p-value of 0.05 of less. If a p-value is less than or equal to 0.05, the results is said to be statistically significant. Be clear on what we are doing here. We are starting from the premise that whatever effect we are trying to estimate does not exist (is zero). We collect our data and do our calculations to get a particular estimate of the effect we are interested in. We compute the p-value for this estimate, which again is the probability that the true effect is zero given the particular estimate that we obtained. If the p-value is small, usually taken to be less than or equal to 0.05, we conclude that our estimated effect is statistically significant. We have rejected the null hypothesis of no effect. If you go on in statistics, you will learn exactly how p-values are computed. They depend on the variability of the population being analyzed. Consider the 200 at-risk students in the summer program experiment. Say that they are all identical, meaning that they will all get the same GPA at the end of four years if they don’t take the summer program. If some do take the summer program, all those who do will get the same GPA, although this GPA will be different if the program does have a non zero effect on GPA. We want to test whether the summer program effect is zero. We run the experiment discussed and get a difference of 0.3. Is this difference statistically significant? The answer is obviously yes. If the true effect were zero, everyone would get the same GPA whether they took the program or not, so the difference would be exactly zero. We in fact got a nonzero estimate, and so we are sure that
|
the true effect is not zero. The p-value would be 0.00. In fact in this case we only need two students, one who took the program and one who did not. If the difference in the two GPAs is not zero, then the summer program has an effect. In this case there would be no need to use intention to treat—there are no selection problems because everyone is identical. Now consider that there is huge variation in the population of 200 regarding what GPA they are going to achieve. Some may turn out to be stars, and some may barely make it through the four years. Whether students take the summer program or not, there will be a huge variation in GPA scores at the end of the four-year period because of the huge variation in the population. We run the experiment and get a difference of 0.3. Is this difference statistically significant? Maybe M21_CASE3826_13_GE_C21.indd 437 17/04/19 4:29 AM 438 PART VI Methodology 21.4 LEARNING OBJECTIVE Understand how regression analysis can be used for both estimation and testing. not if the variation in the population is large. The difference of 0.3 is fairly small, and it could easily be obtained by chance. It just so happened that the particular draw of 100 students led to this outcome, but it may be that a different draw would have resulted in a difference of 0.2. The p-value that is computed for the result of 0.3 would likely be very large, perhaps close to 1.00. The intuition to take from this discussion is that one has more confidence in results obtained from populations with low variation than from those with high variation. To get potentially significant results from a high-variation population, one needs a large sample size. If we had 2,000 at-risk students, gave offers to 1,000, and got a difference of 0.3, this might be significant. When at the end we take the average of the 1,000 GPAs, the individual student characteristics tend to cancel out in a large sample size, and we can have more confidence that the difference of 0.3 is picking up the summer-program effect. When computing p-values, the size of the sample matters as well as the variation in the population. Regression Analysis The most important statistical tool in empirical economics is regression analysis. If you go on in economics you will see applications of regression analysis in microeconomics
|
and macroeconomics. It can be used to forecast the effect of an increase in prices on the quantity of cat food sold in a community or the effects of a stock market decline on household consumption. Here, we provide you with a beginning sense of what regression analysis is all about. There is evidence that the economy has an effect on votes for president in the United States.4 If the economy is doing well at the time of the election, this may have a positive effect on votes for the incumbent-party candidate and vice versa if the economy is doing poorly. This theory suggests that many voters reward or blame the party of the president-in-office for good or bad economic performance while that president is in office. If true, this theory suggests that, all else equal, a president who presides over a strong economy will find his or her political party doing well in the next election. How might we test this theory using regression analysis? We first need some measure of economic performance. The growth rate of the economy is one common measure of economic strength. So we can translate our theory into a more testable form: we postulate that the growth rate of the economy in the year of the election, denoted g, has a positive effect on the incumbent party’s presidential vote share, denoted V. Notice here we have chosen a specific measure of performance—the growth rate—and also a time period—the year of the election. Generally speaking, when we move in economics from a theory to a practical statistical test, we will have some choices to make. In this case we have chosen to measure economic performance by the one-year growth rate. We will look at the way in which the growth rate affects the vote share. In particular, we will assume that V = a + bg (1) If b is positive, this equation says that the growth rate has a positive effect on the vote share, as our theory states. Also, the relationship between V and g is assumed to be linear. If we take a graph with V on the vertical axis and g on the horizontal axis, as in Figure 21.1, the line is straight with intercept a and slope b. The job of regression analysis is to estimate the coefficients a and b and to see in particular if b is positive and if it “matters” in a statistical sense. Consider how we might determine, or estimate, the values for a and b. U.S. presidential elections are held every four years, and there are data
|
on V going back to the beginning of the country. There are also data on g going back many years. If we consider the period beginning in 1916, there have been 26 presidential elections between 1916 and 2016 so we have 26 data points, or observations, on V and g. We can plot these observations in a Figure like 21.1. In the figure we have plotted 10 hypothetical points for illustration. As drawn, the figure shows that there is a positive relationship between the vote share and the growth rate. It also shows, however, that the data points are not all on the line. If equation (1) were exact, all the points would be on the 4See Ray C. Fair, Predicting Presidential Elections and Other Things, 2nd ed. Stanford University Press, 2012, for discussion of this. M21_CASE3826_13_GE_C21.indd 438 17/04/19 4:29 AM V a 0 d1 d2 CHAPTER 21 Critical Thinking about Research 439 slope b ◂◂ FIGURE 21.1 Hypothetical plot of points between the vote share and the growth rate. Observations on V and g g MyLab Economics Concept Check straight line. In fact, in the real world equation (1) is not exact. There are other variables that affect votes for president. Some of these variables include other economic measures, such as perhaps inflation at the time of the election. Vote share may also be affected by foreign policy and personal characteristics and views of the people running for office. As a result, the points in the graph of the vote share on the growth rate are not exactly on the line. The job of regression analysis is to find values of a and b that provide a good fit of the data around the line. Or, in other words, to find the line that best represents the data in the figure. How is fit determined? What do we mean by the best line? This can be seen in Figure 21.1. Draw a particular line with intercept a and slope b. For each data point compute the vertical distance between the point and the line. We have done this for the first two points in the figure, labeled d1 and d2. We do this for all the points, say the 26 observations between 1916 and 2016. Some values of d are positive and some are negative. The larger the distance above or below the line, the worse that particular point fits the line. The distances are usually called “errors” for
|
this reason. The way the fit is determined is first to square each distance. Each squared distance is positive because the square of a negative number is positive. Then we add up all the squared distances, again in our case 26 numbers. Call this sum SUM. The sum is obviously a measure of fit. A small value of SUM means that the points are fairly close to the line, and a large value means they are not. The fact that squared distances are used means that large outliers (distances) are weighted more than small ones in computing SUM. You can think of regression analysis as doing the following, although in practice finding the right line is done more efficiently:5 Try a million different pairs of values a and b, and for each pair compute SUM. This gives us a million values of SUM. Choose the smallest value. The values of a and b that correspond to this smallest value are the best-fitting coefficients—the best-fitting intercept and slope. These estimates are called least squares estimates because they are the estimates that correspond to the smallest sum of the squared distances, or errors. In our theory we focused on the sign of the coefficient of the growth rate: does growth increase vote share? The size of the estimates is often also of interest. If, for example, the estimate of b were 1.0, this tells us that an increase in the growth rate of one percentage point leads to an increase in the vote share of one percentage point. This would be a nontrivial effect of the economy on voting behavior. If the estimate of b were instead 0.01, politicians would worry much less about how a bad economy was going to affect their votes (in practice the estimate is about 0.67). Regression analysis is helpful in letting us test our theories. In our voting example, we are particularly interested in whether b is zero. If b is zero, this says that the growth rate has no effect on votes, and our original theory is not right. To see if we should continue to have confidence in our theory, we need to test whether b is zero. How do we test whether b is zero? Here we go back to what we already know from our discussion of statistical significance and p-values. We first postulate the null hypothesis that b is in fact zero. We then use regression analysis to estimate b, and after this is done we compute the probability (p-value) that we would have obtained this estimate if the truth is that b is zero. If the p-value is low
|
, say less than 0.05, we say that the estimate of b is statistically significant. We reject the null hypothesis that the growth rate does not affect the vote share, and our confidence in the theory that economic performance affects votes is bolstered. 5There are many statistical programs that do this calculation with one simple command, including Microsoft Excel™. least squares estimates Least squares estimates are those that correspond to the smallest sum of squared distances, or errors. M21_CASE3826_13_GE_C21.indd 439 17/04/19 4:29 AM 440 PART VI Methodology Most theories in economics are more complicated than simply one variable affecting another. In our voting example, as noted, inflation may also affect voting behavior. In this case two variables affect V: g and inflation, which will be denoted p. In this case we could write the voting equation as V = a + bg + cp (2) Equation (2) has two variables that explain vote share plus a constant term. There are now three coefficients to estimate rather than two: a, b, and c. With more than one explanatory variable, we cannot draw a graph as we did previously. However, the fitting idea we introduced works the same way when we add variables. Given observations on V, g, and p, you can think of the analysis as trying a million sets of values of a, b, and c and choosing the set that provides the best fit. For each set of three coefficient values, the predicted value of V can be computed for each observation, and the distance for that observation is the difference between the predicted value of V and the actual value of V. We square this distance, do the same for all the observations, and then sum the squared distances. This gives us a value of SUM for the particular set of the three coefficient values. We do this a million times for a million sets of three coefficient values and choose the smallest value of SUM. The coefficient values that correspond to the smallest value of SUM are the least squares estimates of a, b, and c.6 We can also test in a similar manner as discussed whether b and or c are zero. N To conclude, regression analysis is used in many settings. In business, it is used to estimate the size of effects: How much do purchases of a good fall when prices rise? What is the effect of an increase in advertising on car sales? In public policy, magnitudes also matter and can be found using regression analysis:
|
How much more will people use medical care if it is free, and how much will that help their health? How many lives are saved by reducing the speed limit on highways? These are all empirical questions in which regression analysis helps us to get at a magnitude with real consequences. With more data available every day, regression analysis has grown in importance. 6If you go on in economics, you will see that this least squares procedure has to be modified sometimes to account for various statistical problems, but the main goal of trying to find a good fit remains. S U M M A R Y 21.1 SELECTION BIAS p. 429 1. One example of selection bias is survivor bias, where the most fit survive. This makes it difficult to compare young and old age groups. 2. Selection bias can arise if different kinds of people select into different groups, which can bias comparisons of the groups. 21.2 CAUSALITY p. 430 5. Regression discontinuity and difference-in-differences methodologies are also used to identify causality in economics. 21.3 STATISTICAL SIGNIFICANCE p. 437 6. An estimated effect is said to be statistically significant if the probability is small of obtaining the particular estimate when in fact the effect is zero. A probability of less than or equal to 5 percent is commonly used. 3. Correlation is not the same as causality. 21.4 REGRESSION ANALYSIS p. 438 4. Random experiments can sometimes be used to estimate causal effects. Intention to treat is sometimes used with random experiments to deal with limited take up in an experiment. 7. Regression analysis is used to estimate coefficients in equations. It is used both to obtain estimates of the magnitude of effects of various economic factors and to test alternative theories correlated, p. 430 difference-in-differences, p. 435 intention to treat, p. 432 least squares estimates, p. 439 p-value, p. 437 regression discontinuity, p. 433 selection bias, p. 429 statistical significance, p. 437 survivor bias, p. 429 MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M21_CASE3826_13_GE_C21.indd 440 17/04/19 4:29 AM CHAPTER 21 Critical Thinking about Research 441 P R O B L E M
|
S All problems are available on MyLab Economics. 21.1 SELECTION BIAS LEARNING OBJECTIVE: Give some examples of studies that might suffer from selection bias. 1.1 Describe the selection bias likely to exist in the following situations: a. A study of 3,600 children in Australia found that the children who slept early and woke up early had a lower BMI as compared to children who slept late and woke up late. Therefore, children who are early sleepers are healthier than children who go to sleep late. b. A study of 1,500 college students found that students who studied in a private school scored better in the first year of college than students who studied in a government-aided school. Therefore, an education from a private school will improve a student’s academic performance. c. A study of 2,500 women in Sweden found that women who play computer games for an hour or more every day have a higher chance to become obese than ones who do not play them at all. Therefore, women who play computer games have a higher tendency to become obese. 1.2 A classic example of selection bias occurred during World War II. During the war, the British were losing many airplanes over enemy territory and therefore decided to add armor plating to their bombers. The armor was not only heavy, but also expensive, so the British decided to only add armor to the most critical areas of the planes, determined by the location of bullet holes in returning aircraft. The areas most commonly marked by bullet holes were the wings, the nose, and the tail. Before the plan was implemented, Austrian economist Abraham Wald reviewed the data and claimed that the British plan was just the opposite of what was needed, and the armor should be added to the only areas not designated for armor by the British plan: the body and the rudder. The British followed Wald’s recommendation and as a result, many fewer planes were shot down. Explain the selection bias in the original British plan. 21.2 CAUSALITY 2.2 [Related to the Economics in Practice on p. 433] In a randomized one-year trial of 100 elm trees with Dutch Elm Disease, 50 are slated to receive only a fungicide treatment (we will call this Group A), and the other 50 are slated to receive the fungicide treatment and an additional insecticide treatment six months later (we will call this Group B). Assume that the insecticide treatment is ineffective in curing Dutch Elm Disease, so on average, the same
|
proportion of trees in each group will die of the disease. In Group B, five of the 50 trees die in the sixmonth period leading up to the insecticide treatment. Of the 45 trees left, five die in the six months following the insecticide treatment. Since we know the insecticide treatment is ineffective, the trees in Group A will, on average, suffer the same fate as those in Group B, with five trees dying in the first six months and another five dying in the second six months. a. For Group A, what is the rate of death due to Dutch Elm Disease? b. If we limit the analysis in Group B to only those trees which received the insecticide treatment, what is the rate of death due to Dutch Elm Disease? c. What do your answers to parts a and b suggest regarding the reduction in deaths from Dutch Elm Disease due to the addition of the insecticide treatment? d. Knowing what you do about the effectiveness of the insecticide treatment, what is the problem with this analysis? How would applying the intention-to-treat method verify your answer? 2.3 [Related to the Economics in Practice on p. 434] In 1991, economists Joshua D. Angrist and Alan B. Krueger published a study on the correlation between date of birth and years of schooling. The premise was that the actual amount of time an average person spends in school is tied to the time of year in which people are born. Suppose in the city of Gotham, school is mandatory for all children, and students must be six years old by August 31 in order to enter first grade. By law, students must stay in school until they are 16 years old, at which time they can drop out if they choose. How would you use regression discontinuity to evaluate if a person’s date of birth correlates to the years of schooling the person attains in Gotham? LEARNING OBJECTIVE: Understand the difference between correlation and causation. 2.1 Identify each of the following scenarios as examples of causation, positive correlation, and/or negative correlation, and explain your answers. a. Most students who subscribe to online courses tend to perform better academically. b. Most people who are doctors have poor handwriting. c. Most people who live alone are good cooks. d. Most girls who walk to school tend to be healthier than other girls. e. The higher a student’s grades, the less likely he will travel in public buses. 2.4 [Related
|
to the Economics in Practice on p. 436] The neighboring towns of East Magoo and West Magoo are divided by the Quincy River. The towns are similar in geographic size and population. The homes in both towns are powered entirely by electricity provided by Backus County Power and Light, which charges a standardized rate of $0.10 per kilowatt hour (kWh). Both East Magoo and West Magoo add on an additional $0.02 per kilowatt hour as an energy use tax. As a way to increase revenue, the mayor of East Magoo persuaded the town council to double the energy use tax on all residents, effective January 1, 2018. The average monthly energy usage per home is listed in the table below. Use the difference-in-differences method to estimate the effect of the increase in the energy use tax MyLab Economics Visit www.pearson.com/mylab/economics to complete these exercises online and get instant feedback. Exercises that update with real-time data are marked with. M21_CASE3826_13_GE_C21.indd 441 17/04/19 4:29 AM 442 PART VI Methodology on the average monthly amount of energy used per home in East Magoo. Average Monthly Energy Use per Home Scenario 4: In a study to see if Coldplay’s fans or Metallica’s fans spend more, 20 people from each fanbase are surveyed before they enter their respective stadiums for the concerts. Town East Magoo West Magoo 2017 1,775 kWh 1,815 kWh 2018 21.4 REGRESSION ANALYSIS 1,917 kWh 2,033 kWh LEARNING OBJECTIVE: Understand how regression analysis can be used for both estimation and testing. 2.5 Novigrad city has elected a mayor who wishes to enact a new health initiative. She imposes a 10 percent tax on cigarettes to deter people from smoking. After one year, tax revenues from cigarettes increase by 5 percent. Does this mean that the tax increase did not discourage smoking? Explain. 21.3 STATISTICAL SIGNIFICANCE LEARNING OBJECTIVE: Understand how researchers decide whether their results are meaningful. 3.1 Of the following four scenarios, which survey results are likely to be the most statistically significant and which are likely to be the least statistically significant? Explain your answer. Scenario 1: 1,000 students are surveyed after completing six months of a basic French language course to see if they
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.