The Production Function and Changes in Productivity in the European Union

GDP growth is a summary indicator of the economy's capacity to generate real growth and eliminate cyclical slowdowns. To policymakers, the severity of the global financial crisis and the subsequent emergence of the European debt crisis clearly indicate the importance of the production function approach to illustrate policy options during boom-bust episodes.

¹More details about the production function methodology used by the EU are available at Francesca D'Auria et al., “The Production Function Methodology for Calculating Potential Growth Rates & Output Gaps,” Economic Paper No. 420, European Commission, Directorate-General for Economic and Financial Affairs Publications, 2010, Belgium, http://ec.europa.eu/economy_finance/publications/ economic_paper/2010/pdf/ecp420_en.pdf.

(continued)

FIGURE 3.6

The effect of a beneficial supply shock on labor demand

A beneficial supply shock that raises the MPN at every level of labor shifts the MPN curve upward and to the right. Because the labor demand curve is identi­cal to the MPN curve, the labor demand curve shifts upward and to the right from ND¹ to ND 2. For any real wage, firms demand more labor after a beneficial supply shock.

¹⁸For a detailed review of research on labor supply, see the article by Richard Blundell and Thomas MaCurdy, "Labor Supply: A Review of Alternative Approaches," in O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, volume 3 (Amsterdam: North-Holland, 1999), pp. 1559-1695. For more information about key areas of recent research, see the articles by Michael Keane and Richard Rogerson, "Micro and Macro Labor Supply Elasticities: A Reassessment of Conventional Wisdom," Journal of Economic Literature, June 2012, pp.

Equilibrium in the labor market requires that the aggregate quantity of labor demanded equal the aggregate quantity of labor supplied. The basic supply­demand model of the labor market introduced here (called the classical model of the labor market) is based on the assumption that the real wage adjusts reasonably quickly to equate labor supply and labor demand. Thus, if labor supply is less than labor demand, firms competing for scarce workers bid up the real wage, whereas if many workers are competing for relatively few jobs, the real wage will tend to fall. Labor market equilibrium is represented graphically by the intersection of the aggregate labor demand curve and the aggregate labor supply curve at point E in Figure 3.9. The equilibrium level of employment, achieved after the complete adjustment of wages and prices, is known as the full-employment level of employment, N. The corresponding market-clearing real wage is W.

Factors that shift either the aggregate labor demand curve or the aggregate labor supply curve affect both the equilibrium real wage and the full-employment level of employment. An example of such a factor is a temporary adverse supply shock. A temporary adverse supply shock—because of, say, a spell of unusually bad weather—decreases the marginal product of labor at every level of employment. As Figure 3.10 shows, this decrease causes the labor demand curve to shift to the left, from ND¹ to ND². Because the supply shock is temporary, however, it is not expected to affect future marginal products or the future real wage, so the labor supply curve

Effects of a temporary adverse supply shock on the labor market

An adverse supply shock that lowers the marginal product of labor (see Fig.

Unemployment During the European Debt Crisis

Unemployment is considered a lagging indicator, meaning it takes two or three quarters to adapt to changes in the business cycle. This is why it might continue to rise even after the economy has started to recover from a recession.

The impact of the 2007-2009 recession extended beyond the United States and was felt throughout Europe in GDP growth and employment. Greece, Spain, and Italy were among the countries that were most severely affected, and they found it increasingly difficult to continue to bail out financial institutions, refinance gov­ernment deficits, and repay sovereign debt. This led to the European debt crisis as the fiscal positions and debt levels of a number of eurozone countries became unsustainable. These nations had no choice but to seek loans and bailouts from the European Central Bank (ECB), the International Monetary Fund (IMF), and the European Financial Stability Facility (EFSF).²⁶ As part of the bailout packages they received, these countries had to adopt a number of austerity measures by cutting public spending and increasing taxes. This caused further economic contraction, investment slowdown, and unemployment.

As Figure 3.13 shows, unemployment rates started to increase as each of these two crises hit the euro area and continued to increase till the end of the later crisis.

A study by economists from the European Central Bank (ECB) and the na­tional central banks of the EU countries explains this trend.²⁷ During the 2007-2009 recession, most employers chose to steeply cut wages.

As the affected economies started to show signs of recovery, firms responded by hiring more people rather than increasing the number hours worked. As a re­sult, governments of these nations began reverting to stricter labor legislation, with Greece and Spain increasing the minimum wage in 2019.

²⁶The EFSF was a temporary crisis resolution measure created by the EU in 2010 following the sovereign debt crisis. It provided financial assistance to the affected nations.

²⁷Thomas Y. Matha et al., “Shocks and Labour Cost Adjustment: Evidence from a Survey of European Firms,” 2019, ECB Working Paper Series, No. 2269, European Central Bank, April 2019.

⁶The argument was first advanced by the nineteenth-century economist David Ricardo, although he expressed some reservations about its applicability to real-world situations. The word equivalence refers to the idea that, if Ricardian equivalence is true, taxes and government borrowing have equiva­lent effects on the economy.

Measuring the Effects of Taxes on Investment

Does the effective tax rate significantly affect investment patterns? Determining the empirical relationship between tax rates and investment isn't easy. One prob­lem is that the factors other than taxes that affect the desired capital stock—such as the expected future marginal product of capital and real interest rates—are always changing, making it difficult to isolate the "pure" effects of tax changes. Another problem is that changes in the tax code don't happen randomly but reflect the government's assessment of economic conditions. For example, Congress is likely to reduce taxes on investment when investment spending is expected to be unusu­ally low so as to boost spending.

An interesting study that attempted to solve both of these problems was carried out by Jason Cummins of Brevan Howard, Inc., R. Glenn Hubbard of Columbia University, and Kevin Hassett of the American Enterprise Institute.¹¹

¹¹"A Reconsideration of Investment Behavior Using Tax Reforms as Natural Experiments," Brookings Papers on Economic Activity, 1994:2, pp. 1-59. For a survey of work on taxation and investment, see Kevin Hassett and R. Glenn Hubbard, "Tax Policy and Investment," in A. J. Auerbach, ed., Fiscal Policy: Lessons from Economic Research, Cambridge, MA: MIT Press, 1997.

to an increased tax on firm revenue and a rise in the tax-adjusted user cost of capi­tal. Thus, all else being equal, an increase in the effective tax rate lowers the desired capital stock.

Table 4.2 shows effective tax rates on capital for countries in the Organisation for Economic Co-operation and Development (OECD) in 2019. The effective tax rate on capital ranges from -42.4% in Italy to 23.9% in Chile. In the table, the effec­tive tax rate is -2.7% for the United States because of a change in the tax code that allows firms to fully expense the purchase of some types of capital immediately rather than to use depreciation allowances that are spread out over a number of years.¹² Also shown are the ratios of gross investment to GDP in each of these countries. Because an increase in the effective tax rate on capital increases the tax- adjusted user cost of capital, we would expect countries with high effective tax rates on capital to have low rates of investment, all else being equal. However, all else is not equal in the various countries in Table 4.2, and, in fact, there is little sta­tistical relationship between the effective tax rate and investment in this group of countries. In the Application "Measuring the Effects of Taxes on Investment," we describe a successful approach to measuring the impact of taxes on investment.

¹²In 2017, the United States enacted a tax cut called the Tax Cut and Jobs Act, which decreased the statutory corporate tax rate to 21% to 35% and allowed for immediate expensing of expenditures to purchase some types of new capital.

¹⁴Why is investment positive when q is less than one? Even when q for the aggregate economy is less than one, many firms will have q greater than one and will undertake gross investment.

The budget constraint, represented graphically as the budget line, shows the com­binations of current and future consumption available to Penelope. To determine which of the many possible consumption combinations Penelope will choose, we need to know something about Penelope's preferences for current versus future consumption.

Economists use the term utility to describe the satisfaction or well-being of an individual. Preferences about current versus future consumption are summarized by how much utility a consumer obtains from each combination of current and future consumption. We can graphically represent Penelope's preferences for current versus future consumption through indifference curves, which represent all combinations of current and future consumption that yield the same level of utility. Because Penelope is equally happy with all consumption combinations on an indifference curve, she doesn't care (that is, she is indifferent to) which combination she actually gets. Figure 4.A.2 shows two of Penelope's indifference curves. Because the consumption combinations corresponding to points X, Y, and Z all are on the same indifference curve, IC^{* * 1}, Penelope would obtain the same level of utility at X, Y, and Z.

Indifference curves have three important properties, each of which has an eco­nomic interpretation and each of which appears in Fig. 4.A.2:

1. Indifference curves slope downward from left to right. To understand why, let's suppose that Penelope has selected the consumption combination at point Y, where c = 45,000 and c^f = 45,000.⁶ Now suppose that Penelope must reduce her

⁵The present value of current wealth is simply current wealth and the present value of current income is simply current income.

⁶Point Y lies below Penelope's budget line, shown in Fig. 4.A.1, which means that not only could Penelope afford this consumption combination, but she would have resources left over at the end of the future period. Unless she wants to leave a bequest, she would not actually choose such a combi­nation for the resources shown in Fig. 4.A.1.

We are now ready to investigate the economic forces that determine international trade and borrowing. In the remainder of this chapter we demonstrate that a coun­try's current account balance and foreign lending are closely linked to its domestic spending and production decisions. Understanding these links first requires devel­oping the open-economy version of the goods market equilibrium condition.

In Chapter 4 we derived the goods market equilibrium condition for a closed economy. We showed that this condition can be expressed either as desired national saving equals desired investment or, equivalently, as the aggregate supply of goods equals the aggregate demand for goods. With some modification, we can use these same two conditions to describe goods market equilibrium in an open economy.

Let's begin with the open-economy version of the condition that desired national saving equals desired investment. In Chapter 2 we derived the national income accounting identity (Eq. 2.9):

Recent Trends in the U.S. Current Account Deficit

The United States has a very large current account deficit, as discussed in Section 5.1, "Balance of Payments Accounting." But the current account deficit shrank from 2007 to 2009, thanks to the U.S. recession, which reduced demand for imported goods.

Figure 5.7 shows how the current account balance has changed since 1960. The figure shows that the current account balance was generally positive in the 1960s, bounced up and down around zero in the 1970s, and then fell sharply to about -3% of GDP by the mid 1980s. In the second half of the 1980s, the current account balance improved, becoming positive again in 1991. From 1991 to 2005, it deteriorated markedly, falling to -6.2% of GDP by the end of 2005. The housing downturn of 2006 and the recession from 2007 to 2009 helped the current account balance improve to about -5% of GDP by mid 2008, as U.S. consumers bought fewer imported goods and services than before. When the financial crisis hit in late 2008, U.S. imports fell sharply and the current account balance improved to about -2.5% of GDP by mid 2009. As the U.S. economy improved slowly from 2009 to 2018, the current account balance remained higher than it was before the financial crisis at -2% to -3%. In the pandemic recession of 2020, the current account balance got even more negative, falling below -3%.

A major reason for the increase in the current account deficit from 1998 to 2005 was a change in the saving behavior of developing countries.⁹ Developing nations that had borrowed from abroad in past decades began lending abroad, especially to the United States. In part, this change in behavior occurred as a result of past financial crises, such as Mexico experienced in 1994 and the countries of East Asia experienced in 1997-1998. Those crises were exacerbated by the fact that those countries had borrowed substantially from foreign countries. To pre­vent such crises from happening again, the countries increased their saving, built

⁹For an analysis of this view, see Ben S. Bernanke, "The Global Saving Glut and the U.S. Current Account Deficit," Sandridge Lecture, Virginia Association of Economists, Richmond, Virginia, March 10, 2005. Available at wwwfederalreserve.gov/boarddocs/speeches/2005. d)

The Twin Deficits

The relationship between the U.S. government budget and the U.S. current account for the period 1960-2021 is illustrated in Figure 5.9. This figure shows government purchases and net government income (taxes less transfers and interest) for the Federal government alone as well as for the combined Federal, state, and local gov­ernments, all measured as a percentage of GDP. Our discussion of the current account balance as the excess of national saving over investment leads us to focus on the broadest level of government, which includes state and local government in addition to the Federal government. We also present data on the purchases and net income of the Federal government alone because Federal budget deficits and surpluses are often the focus of public attention. In addition, as shown in Fig. 5.9, much of the movement in purchases and net income of the combined Federal, state, and local gov­ernment reflects movement in the corresponding components of the Federal budget.

The excess of government purchases over net income is the government budget deficit, shown in beige.¹³ Negative values of the current account balance indicate a current account deficit, also shown in beige. Large deficits—both government and current account—occurred during the 1980s and much of the 1990s.

¹³Government purchases are current expenditures minus transfer payments and interest. Thus gov­ernment investment is not included in government purchases. The budget deficit is the current defi­cit; see Chapter 15 for a further discussion of this concept.

Waves of Productivity Growth over Time

We have seen in Table 6.3 that productivity growth in the United States has aver­aged a bit under 1% per year from 1929 to 2020, but in waves. Productivity grew faster than average from 1948 to 1973, but then declined from 1973 to 1982. After 1982, productivity growth was slightly below 1% again, though the period in the late 1990s was marked by rapid growth of productivity.

Waves in productivity growth have been going on for some time.⁵ In the early 1800s, the development of water power, textiles, and iron led to rapid improve­ments in the standard of living. Another wave of innovation came in the mid- to late 1800s in the form of steel, steam, and railroads. The next wave began in the early 1900s with the internal combustion engine, electrification, and the use of chemicals. The 1950s and 1960s brought commercial aviation, electronics, and petrochemicals. Finally, the late 1990s and early 2000s saw the development of the Internet and cell phones. Today, we think that artificial intelligence and self­driving cars are likely to be part of the future, but studies of past innovations suggest that it might take a couple of decades before productivity growth rises significantly as a result of those innovations.

Will improved productivity and the next wave of innovation destroy jobs and lead to poverty? Some analysts think that self-driving cars and robotics could replace much human work activity. As a result, some have called for governments to provide a guaranteed income for workers who are displaced and might never work again. As the labor-market theory in Chapter 3 notes, improved technology normally increases the marginal product of labor. For example, artificial-intelligence programs are being developed that can help doctors make more accurate diagno­ses, and robots on the assembly line allow a given number of workers to produce more cars with fewer defects. Increases in the marginal product of labor brought about by better technologies imply that labor demand and wages should rise, not fall. Indeed, as Joel Mokyr of Northwestern University, Chris Vickers of Auburn University, and Nicolas L. Ziebarth of the University of Iowa note, people have worried about technology disrupting their lives for generations.⁶ Generally, pre­dictions about the demise of the labor force and falling wages have been incorrect and misguided. Could this time be different?

⁵For an early discussion, see Nikolai D. Kondratieff, “The Long Waves in Economic Life,” Review of Economics and Statistics, November 1935, pp. 105-115.

⁶"The History of Technological Anxiety and the Future of Economic Growth: Is This Time Different?” Journal of Economic Perspectives, Summer 2015, pp. 31-50.

The M2 Monetary Aggregate. Everything in M1 plus other assets that are somewhat less "moneylike" compose M2. The main additional assets in M2 include small-denomination (under $100,000) time deposits and noninstitutional holdings of money-market mutual funds (MMMFs). Both of these accounts have limits on the number of checks or other transfers that can be written each month. Time deposits are interest-bearing deposits with a fixed term (early withdrawal usually involves a penalty). As mentioned, MMMFs invest their shareholders' funds in short-term securities, pay market-based interest rates, and allow holders to write a limited number of checks.

The Money Supply. The money supply is the amount of money available in an economy.⁴ In modern economies the money supply is determined by the central bank—in the United States, the Federal Reserve System.

A detailed explanation of how central banks control the money supply raises issues that would take us too far afield at this point, so we defer that discussion to Chapter 14. To grasp the basic idea, however, let's consider the simple hypotheti­cal situation in which the only form of money is currency. In this case, to increase the money supply the central bank needs only to increase the amount of currency in circulation. How can it do so?

One way—which is close to what happens in practice—is for the central bank to use newly minted currency to buy financial assets, such as government bonds, from the public. In making this swap, the public increases its holdings of money, and the amount of money in circulation rises. When the central bank uses money to purchase government bonds from the public, thus raising the money supply, it is said to have conducted an open-market purchase.

To reduce the money supply, the central bank can make this trade in reverse, selling government bonds that it holds to the public in exchange for currency. After the central bank removes this currency from circulation, the money supply

⁴The terms money supply and money stock are used interchangeably.

¹⁶To derive velocity under the quantity theory, we must assume that nominal money demand, M^d, equals the actual money stock, M, an assumption that we justify later in the chapter. Under this assumption, you should verify that V = 1∣ k.

¹⁷We relax this assumption in Part 3 when we discuss short-run economic fluctuations.

⁴Although a drop in output, Y, obviously reduces the quantity of goods supplied, it also reduces the quantity of goods demanded. The reason is that a drop in output is also a drop in income, which reduces desired consumption. However, although a drop in output of one dollar reduces the supply of output by one dollar, a drop in income of one dollar reduces desired consumption, C^d, by less than one dollar (that is, the marginal propensity to consume, defined in Chapter 4, is less than 1). Thus a drop in output, Y, reduces goods supplied more than goods demanded and therefore reduces the excess supply of goods.

Another type of shock that can be a source of business cycles in the classical model is a change in fiscal policy, such as an increase or decrease in real government purchases of goods and services.¹³ Examples of shocks to government purchases include military buildups and the initiation of large road-building or other public

¹³Another important example of a change in fiscal policy is a change in the structure of the tax code. Classical economists argue that the greatest effects of tax changes are those that affect people's incen­tives to work, save, and invest, and thus affect full-employment output. However, because most classi­cal economists accept the Ricardian equivalence proposition, they wouldn't expect lump-sum changes in taxes without accompanying changes in government purchases to have much effect on the economy.

Inflation and unemployment in the United States, 1970-2021

The figure shows the combinations of inflation and unemployment experienced in the United States each year since 1970. Unlike during the 1960s (see Fig. 12.1), after 1970 a clear nega­tive relationship between inflation and unemploy­ment in the United States didn't seem to exist.

Source: Same as for Fig. 12.1.

We have addressed two of the questions about the Phillips curve raised earlier in the chapter—the questions of why the Phillips curve was observed in historical data and why it seemed to shift after 1970. We still must answer the third question: Can the Phillips curve be thought of as a "menu" of inflation-unemployment combinations from which policymakers can choose? For example, can policymak­ers reduce the unemployment rate by increasing the rate of inflation (moving up and to the left along the Phillips curve)?

According to the expectations-augmented Phillips curve, unemployment will fall below the natural rate only when inflation is unanticipated. So the question becomes: Can macroeconomic policy be used systematically to create unanticipated inflation?

Classical and Keynesian economists disagree on the answer to this question. Classicals argue that wages and prices adjust quickly in response to new economic information, including information about changes in government policies. Furthermore, classicals believe that people have rational expectations, meaning that they make intelligent forecasts of future policy changes. Because prices and price-level expectations respond quickly to new information, the government can't keep actual inflation above expected inflation—as would be needed to drive unemployment below the natural rate—except perhaps for a very short time. According to classicals, policies (such as more rapid monetary expansion) that increase the growth rate of aggregate demand act primarily to raise actual and expected inflation and so do not lead to a sustained reduction in unemployment. Because any systematic attempt to affect the unemployment rate will be thwarted by the rapid adjustment of inflation expectations, classicals conclude that the Phillips curve does not represent a usable trade-off for policymakers. ("In Touch with Data and Research: The Lucas Critique," explores a general lesson for policymakers of the shifting Phillips curve.)

In contrast, Keynesians contend that policymakers do have some ability—in the short run, at least—to create unanticipated inflation and thus to bring unem­ployment below the natural rate.⁹ Although many Keynesians accept the notion

⁹As we discussed in Chapter 11, Keynesian economists also believe that macroeconomic policy can be used to return the unemployment rate to its natural level if the economy starts out in a recession or a boom.

⁸There are actually two different discount rates. The primary credit discount rate applies to banks in good condition. The secondary credit discount rate is higher than the primary credit discount rate and is charged to banks that are in worse condition; such banks will also be subject to careful supervision by the Fed.

Another method a central bank can use to get around problems caused by the zero lower bound is to increase its total holdings of assets, known as quantitative easing. Under quantitative easing, the Fed buys bonds in the open market. The sellers of the bonds deposit the proceeds from the sale in their banks, which adds to banks' reserves. Higher bank reserves increase high-powered money and the

¹¹Kevin Peachey, "What Low Interest Rates Mean for You," BBC News, March 5, 2009, http://news.bbc.co.uk/1/hi/business/7924201.stm.

¹²See Federal Reserve issues FOMC statement, 2012 at www.federalreserve.gov/newsevents/pressreleases/ monetary20121212a.htm.

Interest rate targeting

The figure shows an economy that is buffeted by nominal shocks. Changes in the money supply or money demand cause the LM curve to shift between LM² and LM³ and cause aggregate demand to move erratically between Y₂ and Y₃. A Fed policy of keeping the real inter­est rate at r₁, by raising the monetary base when­ever the interest rate exceeds r₁ and lowering the base whenever the interest rate falls below r₁, will keep the economy at full employment at E.

Inflation Targeting

By the early 1980s, most countries in the Organization for Economic Co-operation and Development (OECD) were already announcing some form of money or credit target to convince markets of their commitment to control inflation. However, the focus was on "intermediate targets" such as the quantity of money or credit, rather than inflation. Although price stability is acknowledged as the long-run goal of monetary policy, there is no consensus on how monetary policy can be conducted to achieve it. A widely used monetary policy regime in recent times is inflation targeting (IT).³² A credible inflation target becomes a nominal anchor, providing conditions that make the price level uniquely determined. This is necessary for price stability and helps tie down inflation expectations directly.

By committing to inflation targeting, countries make an explicit commitment to meet a specified inflation rate (or range) within a certain time frame. Specific elements include (a) regular public announcements of medium-term numerical targets, or ranges, for inflation, with a reduced role for intermediate targets such as money growth; (b) an institutional commitment to price stability as the primary, long-run monetary policy goal and to achievement of the inflation target; and (c) accountability of the central bank for meeting the target.

New Zealand became the first country to formally adopt inflation targeting as we now know it when The Reserve Bank of New Zealand Act, 1989, came into effect in February 1990. The policy arose from widespread dissatisfaction with the economy's performance over the decade, especially in 1984, when inflation averaged over 10% and economic growth rate was abysmal. Since 1990, several central banks from every

³²See B.S. Bernanke, and ES. Mishkin, "Inflation Targeting: A New Framework for Monetary Policy?" Journal of Economic Perspectives 11, no. 2, 1997, pp. 97-116 for a comprehensive discussion.

(continued)

Another strategy for improving a central bank's credibility is to increase the independence of the central bank from the other parts of the government—for example, by limiting the legal ability of the executive and legislative branches to interfere in monetary policy decisions. The rationale is that a more independent central bank will be less subject to short-term political pressures to try to expand output and employment (say, before an election) and will be more strongly com­mitted to maintaining a low long-run inflation rate. Because the public will recog­nize that an independent central bank is less subject to political pressures, announcements made by the central bank should be more credible.

³⁴The point that appointing a tough central banker may improve central bank credibility was made by Kenneth Rogoff, "The Optimal Degree of Commitment to an Intermediate Monetary Target,” Quarterly Journal of Economics, November 1985, pp. 1169-1189. Keynesians might argue that Volcker was credible but that long-lived stickiness in wages and prices led unemployment to increase in 1981-1982 anyway.

³²"The Monetary Dynamics of Hyperinflation," in Milton Friedman, ed., Studies in the Quantity Theory of Money, Chicago: University of Chicago Press, 1956.