Great Depression Economics 101
Has another Great Depression begun? Consider what happened to the U.S. economy in March 2020. In the last week of the month, a record 6.6 Million Americans sought unemployment benefits. The March report from the Department of Labor showed job losses during the month to have been 701,000 jobs. The unemployment rate for March rose to 4.4% from 3.5% in February. This was the largest one-month increase in the rate since January 1975. The Congressional Budget Office forecast that during the second quarter of 2020, the unemployment rate would exceed 10%. Notably, a monthly unemployment rate of 10.8%, which occurred in 1982, represents the highest monthly unemployment rate recorded since 1948, when record keeping began. In mid-month the Federal Reserve announced that to provide sufficient economic stimulus it would reduce interest rates to zero, by purchasing hundreds of billions of dollars of bonds. At the end of the month, President Trump signed a record stimulus bill worth $2 trillion after it was passed by Congress.
Keynes’ General Theory and Economic Depression
John Maynard Keynes’ book The General Theory of Interest, Employment, and Money is one of the classic works of the twentieth century. Keynes published his book in 1936 during the midst of the Great Depression, with his book explaining the cause of the depression as well as measures to address the contraction.
Because the risk is high that our world has just entered Great Depression II, it seems worthwhile to remind ourselves what Keynesian theory has to tell us about the dynamic of an economy that enters a depression, including possible policy responses.
Unlike Great Depression I, which was induced by a large negative shock to aggregate demand, our (possibly) unfolding Great Depression II was induced by a negative labor supply shock stemming from a pandemic. In this post, I discuss why the Keynesian framework is robust enough to accommodate the fact that the COVID-19 pandemic impact on the economy reflects the combination of a major negative labor supply shock as well as an associated negative aggregate demand shock.
Below I describe an economic depression scenario generated by a supply shock, viewed through the lens of a standard Keynesian macroeconomic equilibrium model for a closed economy.
Purpose of This Post
I have written this post as a tutorial for readers who have some background in textbook macroeconomics, and have an interest in understanding how to analyze an economic depression using Keynes’ framework. My reason for doing so is to provide a structure for systematic, reasoned thought about how to deal with what might well be a prolonged pandemic-induced economic contraction, with major social, political, and health challenges.
In the pandemic-induced depression scenario, a sudden negative supply shock reduces the labor supply (in person-hours), thereby causing sharp decreases in employment, output, and income. Consider these questions: As the economy contracts, what happens to corporate investment, prices, interest rates, unemployment, fiscal policy, and monetary policy? How does the Keynesian framework help us consider these questions as well as help us assess policy options for recovery?
General Structure of Keynesian Macroeconomic Model
In the textbook Keynesian macroeconomic model, macro-equilibrium for the economy is depicted as the intersection of two curves, one for aggregate supply and the other for aggregate demand. Aggregate supply corresponds to gross domestic product (GDP). Aggregate demand is the sum of three types of claims on GDP: consumption demand by consumers (C), corporate investment demand (I), and demand by government (G). There is also net demand from international trade, the difference between exports and imports; however, for simplicity I will ignore this component in the discussion below.
As a whole, the economy requires an amount of income (Y) to be able to purchase its output (GDP) exactly. This means that Y = GDP. In equilibrium, total aggregate demand (C+I+G) must equal total supply (GDP or Y). The aggregate demand and aggregate supply curves depict aggregate demand and aggregate supply as functions of the price level (P). Overall equilibrium for the economy occurs at the intersection between supply and demand curves.
The Keynesian framework has as its focus the interaction between three broad markets in the economy, namely the product market, the money market, and the labor market. The aggregate demand curve captures most of the features of the product and money markets, while the aggregate supply curve captures salient features of the labor market.
Figure 1 below is a graphical depiction of a supply-demand equilibrium for the economy. In Figure 1, the income (output) variable (Y) is on the vertical axis and the price level (P) is on the horizontal axis.
The supply curve is upward sloping in P, reflecting the employment and production decisions firms make as they seek to maximize profits. Firms only increase production if the increased revenue at least covers the additional costs of paying workers; and the cost of paying workers depends on the nominal wage rate (W). If firms start out maximizing profits, and prices rise but not nominal wages, then firms will have an incentive to increase output and employment. The aggregate supply curve is increasing in P because when P goes up and the nominal wage (W) is fixed, the real wage (W/P) falls, and therefore the profit maximizing level of output rises. Notice in Figure 1 that the supply curve flattens out at full employment, which means that aggregate supply (Y) is bounded from above.
Along the aggregate demand curve, output Y decreases as a function of price level P. Each value of Y along the aggregate demand curve emerges as an equilibrium condition associated with a Keynesian “IS-LM intersection.” The I in IS refers to corporate investment (I) and the S refers to consumer savings (S). LM refers to investors’ demand for cash or liquid assets (L) and the supply of money in the banking system (M).
Figure 2 illustrates an IS-LM intersection, with IS and LM as curves featuring output and income (Y) on the horizontal axis and the interest rate (i) on the vertical axis.
The IS curve is downward sloping, because firms respond to a higher interest rate (i) by reducing corporate investment (I) which in turn reduces aggregate demand (C+I+G). Keynes emphasized that consumption (C) would positively depend on income (Y). Therefore, a higher interest rate (i) leads to a lower Y associated with the relationship C+I+G = Y.
The LM curve is upward sloping, with a flat portion at the bottom representing what is known as a “liquidity trap:” the trap simply means that the central bank faces a lower limit on its ability to reduce the nominal interest rate. The LM curve is upward sloping to reflect the operation of the money market and banking system. Money used to underlie economic transactions is called transaction demand. The higher is economic activity (Y), the greater the transaction demand. Money is also held by investors as the cash portion of their portfolios. When interest rates rise, investors tend to reduce their cash positions and shift to bonds, in order to achieve higher yields.
Here is what makes the LM curve upward sloping along most of the curve. When economic activity (Y) increases, and the central bank maintains the supply of money (M) to the economy, the increased transaction demand for money will produce upward pressure on interest rates. This is because higher interest rates induce investors to shift some of their cash holdings into bonds, thereby freeing up cash to support the higher transaction activity.
As the price level P increases, the LM curve shifts up. This is because a higher price level also increase the transactions demand for money, with a corresponding need to to induce “bond investors” to shift from cash to bonds. Therefore, a higher price level P leads interest rates to rise, pushing the IS-LM intersection back up the IS curve to lower Y (from reduced investment (I)). Figure 3 illustrates the dynamic just described.
COVID-19 Pandemic Represented as a Negative Supply Shock
The containment measures to combat the COVID-19 pandemic have reduced the effective supply of person-hours to the economy, resulting in a negative supply shock. Such a negative shock manifests itself in the aggregate supply curve of Figure 1, resulting in lower supply Y for every price level P. That downward shift to the supply curve pushes the supply-demand intersection point downward and to the right (along the demand curve). This means a higher equilibrium price level P and lower output level Y.
Notice that a negative supply shock, with unchanged aggregate demand function, reduces equilibrium income, and output, but pushes up the prices of goods and services. This is because profitably filling the excess demand requires firms to face lower real wages; and with nominal wages the same, the only profitable way for increased production to occur is for the price level to increase. The new equilibrium at higher prices requires aggregate demand to decline, and this happens along the original aggregate demand curve when increased prices induce increased interest rates in order to provide cash for the higher transactions activity.
The upward price pressure that accompanies the contraction in output and income will be short lived. This is because firms will react to the decline by reducing corporate investment. Consumers will react to the contraction by cutting back on expenses, for both autonomous reasons, and because their income declines. The model captures the impact of reduced corporate investment and reduced autonomous consumption though a leftward shift in the IS curve.
For a fixed LM curve, the shift in IS will move the equilibrium down the LM curve, reducing both Y and interest rates at least until reaching the liquidity trap, when the central bank can no longer lower interest rates by using expansionary monetary policy. Traditionally, the lower bound on nominal interest rates has been viewed as zero, but in recent years nominal rates have gone negative in some countries, such as Japan. As I mentioned at the outset of this post, in mid-March the Federal Reserve announced that it would use open market operations to bring the U.S. short term rate to zero.
The declines in corporate investment and autonomous consumption push the aggregate demand curve down. What Figure 5 shows is that for any price level P, the leftward shift in IS curve results in lower output Y. Therefore, at the beginning of the contraction, both the aggregate supply and the aggregate demand curve will shift down. See Figure 6.
Notice that in Figure 6, the equilibrium level of the price level P falls. During economic contractions, the possibility of deflation arises.
Policy to Address the Economic Downturn
Governments can respond to negative shocks using a combination of fiscal and monetary policy. Increased government spending and tax cuts will shift the IS curve back to the right, capturing the impact of the $2 trillion stimulus package. (The IS curve might be re-termed the ISG curve, to reflect government fiscal policy in addition to saving and investment.) Expansionary monetary policy shifts the LM curve to the right, subject to the liquidity trap, capturing the impact of the Fed’s policy of aggressively buying bonds. The reason why expansionary monetary policy shifts the LM curve to the right is as follows: The additional liquidity provided by the central bank can be used to fulfill some of the transaction demand for money. For this reason there less upward pressure on interest rates is required to induce investors to shift from cash to bonds, in order to free up cash for transaction purposes. The stimulus from both fiscal policy and monetary policy serve to shift the aggregate demand curve upwards from their depressed states.
To recapitulate: Figure 6 illustrates that relative to an initial situation featuring near full employment, the COVID-19 supply shock generates a dramatic increase in unemployment and dramatic decreases in income and output. In the IS-LM interaction underlying the aggregate demand curve, lower interest rates take the economy into the liquidity trap where monetary stimulus fails to increase aggregate demand. The aggregate demand curve does not shift with further monetary expansion, although investors hold more cash in their portfolios, as the central bank continues to buy bonds. The stimulus package and central bank expansionary policy partially offsets the declines in corporate investment and autonomous consumption.
Economic depression continues through the duration of the supply shock. Figure 6 makes clear that the economy can only fully recover when the pandemic subsides and person hours supplied to the economy return to near its previous level. Neither fiscal nor monetary stimulus will change this state of affairs.
This is not to say that government policy is futile. In Figure 6, fiscal and monetary stimulus can generate some benefit, because the post-shock equilibrium lies along the upward sloping portion of the aggregate supply curve. In addition, redistribution-based safety net policies and jobs programs can produce important benefits. However, the impact of stimulative policies are limited by the containment measures needed to address the spread of the pandemic. Notably, the strength of containment is largely a policy variable, although many individuals will choose to engage in self-isolation.
The $2 trillion stimulus package mentioned above roughly corresponds to 10% of the size of the economy. Figure 6 depicts a situation in which the supply shock leads both income and the price level to fall, with the equilibrium level of income being at its maximum given the containment policy. In this type of situation, although fiscal and monetary stimulus has limited ability to increase real output, stimulus can increase the price level. The challenge for policy makers is to aim for a stimulus package that is neither too weak and allows for deflation, nor too strong and produces inflation.
Moreover, as containment measures are lifted, a restoration of the aggregate supply curve to something like its original position will not necessarily end the depression. This is because the aggregate demand curve also needs to return to something like its original position. In this respect, the Great Depression occurred mostly because of a negative shock to the aggregate demand curve, not the aggregate supply curve.
In other words, for the depression to end, the graphical story portrayed above must play out in reverse.
Inflation and Unemployment
When the economy recovers, and unemployment falls, the price level will begin to rise. According to the model, the price level must go up in order to drive down the real wage rate, as lower real wages are required to induce firms to hire more labor. Eventual resistance to declining real wages will lead workers to negotiate higher nominal wages. A higher nominal wage rate will shift the aggregate supply curve to the right, leading to a new equilibrium featuring lower output and income Y and a higher price level P. If government policy seeks to stimulate the economy to stem the downward pressure on Y, then the aggregate demand curve will shift right, thereby leading to an equilibrium with the original value of Y. However, the new equilibrium will feature a yet higher price level and lower real wage. See Figure 7.
According to the model, the economy will not be able to sustain low unemployment and a high real wage. Therefore, labor has to accept a low real wage in order to achieve low unemployment and stable prices. The attempt to achieve both low unemployment and a high real wage will lead to ever increasing inflation, as workers set nominal wages in anticipation of higher future inflation.
Current corporate investment generates fixed assets for growing the economy. Higher current investment leads to higher future aggregate supply curves. Therefore, lower corporate investment during the depression leads to slower future growth and lower future output.
In future posts, I will discuss empirical issues associated with the evolution of the U.S. economy as it goes through Great Depression II. In doing so, I will intermittently refer to this post, as the theory provides a coherent framework for thinking about how all the various parts of the economy fit together.