I don’t know if I can stomach much more of the posturing in Washington. We know it is all about politics. And Congressman Ryan has yet again said that budget cutting is about “morality”, not economics. The possibility that a sovereign government might be shut-down this weekend because its elected representatives will not extend a self-imposed debt limit just boggles the mind. I’m so fed up I really cannot write about that nonsense any more.
Yet, the insanity has seeped outside the beltway. Even economists—who should know better—are weighing-in in favour of Washington’s budget-cutting. And not just any economists—even those with Bank of Sweden-awarded “Nobels” (mind you, not real Nobel prizes) and assorted other distinctions have come to the support of the craziest Tea Party ideas. Cutting government spending is good for growth! The fiscal stimulus package cost us jobs! Worker pensions caused the fiscal shortfall facing states! Keynes and Roosevelt are responsible for all our current problems! Time to bring back Hoovernomics!
For example, in a Wall Street Journal article this week three Hoover Institute economists (Gary Becker, George Schultz and John Taylor) endorsed Republican efforts to make large federal government budget cuts. (http://www.marketwatch.com/story/time-for-a-budget-game-changer-2011-04-03) I will not address all the arguments made in defence of a “Hooverian” approach to economics. Instead I want to focus on the two main points made. These are arguments that any student of Econ 101 from 1950 up to the present day would have been able to destroy with both argument and evidence. Here are their arguments:
1. “When private investment is high, unemployment is low. In 2006, investment—business fixed investment plus residential investment—as a share of GDP was high, at 17%, and unemployment was low, at 5%. By 2010 private investment as a share of GDP was down to 12%, and unemployment was up to more than 9%. In the year 2000, investment as a share of GDP was 17% while unemployment averaged around 4%. This is a regular pattern.”
2. “In contrast, higher government spending is not associated with lower unemployment. For example, when government purchases of goods and services came down as a share of GDP in the 1990s, unemployment didn’t rise. In fact it fell, and the higher level of government purchases as a share of GDP since 2000 has clearly not been associated with lower unemployment.”
They then supply a graph showing investment and government spending as a share of GDP to demonstrate these two points. Based on that data, these Hooverians argue that the solution is to cut federal spending and then to hold its growth rate below that of GDP. This will allow the share of government spending to fall—while economic growth will let tax revenues rise a bit faster so that the budget will move toward balance. (Indeed, tax revenue does tend to grow much faster than GDP in a boom—increasing as a per cent of GDP–which is how we got the Clinton-era budget surpluses.)
By framing their argument in terms of ratios to GDP, the authors provide a misleading characterization of cause and effect. It is true that high investment spending tends to increase GDP while lowering unemployment—that is the Keynesian “multiplier” at work. High growth of GDP, in turn, lowers the ratio of government spending to GDP so that we will observe a correlation between falling unemployment and a falling government share of GDP—but that is a correlation of no causal significance. When an investment boom collapses—as it did in 2006-2007—GDP growth then falls and the government share of a smaller GDP will rise. Our Hooverians interpret that as “proof” that a rising government share does not help to fight unemployment.
In fact, however, relatively stable government spending over a cycle helps to cushion a private sector “bust”. While it is hard to prove the counterfactual—how bad would things have been without sustained government spending?—it is hard to believe their argument that a loss of 8% of GDP due to reduction of private spending would not have led to a much deeper recession (or depression) without the stabilizing force of our government spending. Simply arguing that we did not get recovery in spite of the Obama stimulus ($800 billion over two years, or well under half the loss of private sector spending) will not do—it does not tell us how high unemployment would have gone in the absence of stimulus.
Let us take a look at the components of GDP over the past two decades. Recall from your Econ 101 course that the aggregate measure of a nation’s output of goods and services (GDP) is equal to the sum of consumption, private investment, government purchases, and net exports (for the US that is of course negative). We can further divide investment into residential (housing) and nonresidential (investment by firms). Finally, we can divide government spending between federal government and state and local government. The following chart graphs the domestic components of GDP (net imports are left out), indexing each component to 100 in 1990. (This makes the scale easier to show in the graph, and simplifies comparison of growth by component. For example, if consumption spending doubles between 1990 and 2000, its index increases from 100 to 200.)
What we see in this graph is that the slowest growing component over the two decades was federal government spending—it actually did not grow much until the term of President George W. Bush. (A substantial portion of federal government growth since 2000 can be attributed to our multiple wars, as well as to domestic spending on security in the aftermath of 9-11.) By 2010, federal government spending was just over 2.3 times bigger (in nominal terms) than its spending in 1990. This graph certainly does not appear to show that federal government spending has been growing so fast that it is “crowding out” private spending on investment. Indeed, it was only after President Bush’s spending increases for foreign wars and domestic security that we see obvious outsized booms in residential and nonresidential investment. There is no evidence that government purchases crowded out private investment spending.
Private consumption as well as state and local government spending grew steadily, increasing by about 267% before the deep recession led to some retrenchment. Note that the fiscal crisis now facing cities and states will lead to continuing cutbacks in state and local government spending—the “perfect fiscal storm” I wrote about last time. And while consumption appears to be recovering by 2010, the jury is still out. Still, the overriding picture one gets is that consumption as well as state and local government spending have been growing relatively steadily, and at a pace considerably faster than growth of federal government spending.
By contrast, residential investment boomed in the real estate bubble, growing 350% by 2005. It then collapsed so that it stood at an index of just 150 in 2010 (50 per cent higher than in 1990). Nonresidential investment shows a clear cyclical nature, and it too collapsed in the aftermath of the global financial collapse. Viewed in this light, it is not at all surprising that when total investment (residential plus nonresidential) is growing rapidly, unemployment tends to fall; but when investment spending collapses we lose jobs at a stupendous pace. In a small government economy, it is investment that dominates, through the Keynesian multiplier: when firms and households are optimistic, we get residential and nonresidential investment, growth, and jobs; when they are pessimistic we get recession and unemployment.
This has long been the concern of Keynesian economists: investment by its very nature is highly cyclical, subject to what J.M. Keynes called “whirlwinds of optimism and pessimism”. That is not all bad. J. Schumpeter referred to the “creative destruction” that makes capitalism dynamic—waves of innovation generate new investment, wiping out firms that get left behind. But if an entire economy is whipped about by unstable investment, we oscillate between the extremes of boom and bust. That is why we need some spending that is more stable—better yet, we need a source of spending that can act in a countercyclical manner to offset the swings of investment.
And that is precisely what we created in the aftermath of the Great Depression. First, we grew the federal government—from about 3% of GDP in 1929 to above 20% after WWII. As Hyman Minsky used to say, government needs to be at least as big as investment to ensure it can offset swings of investment spending. As the chart above shows, federal government spending is not subject to the wild swings that afflict investment, so it helps to stabilise GDP and jobs—if it is big enough. Second, we put in place a variety of federal government programs that help to stabilise household consumption (unemployment benefits, Social Security retirement, and “welfare” for households, firms, and farms). That is, again, reflected in the chart above—even when the financial sector crashed and unemployment exploded, consumption dipped only slightly, thanks in large part to government “transfer” payments like unemployment benefits. Note that transfer payments are not explicitly included in the chart above. Rather, payments like unemployment benefits and Social Security show up in consumption—helping to stabilise it over the course of the cycle.
Our modern Hooverians would like to return to the “good old days” of President Hoover, when the government was smaller and both unwilling and unable to offset the swings of private investment spending. Back then, when investment collapsed unemployment did not go to 9 or 10 per cent, it went all the way to 25 per cent. When people lost their jobs, there was no “welfare” to fall back on. Hence, consumption would collapse—feeding through to lost retail sales, and on to farm products, and thence throughout every corner of the economy. That is why GDP fell in half, and why the Great Depression was so “great”. Government was far too small to stabilise spending and incomes.
Sure, we got the New Deal programs that helped a bit. Unfortunately, President Roosevelt lost his nerve in 1936 in the face of budget deficits. He raised taxes and constrained spending in an attempt to reduce the deficit. In fact, the following crash in 1937 was the sharpest in US economic history—even worse than the 1929 crash. The New Deal programs, by themselves, would never have generated recovery, because they were too small—just like the Obama stimulus. Instead it was WWII that ramped up government, increased budget deficits to 25%, and increased production to the full employment level. The rest, as they say, is history.
Or, at least it was. Until modern Hooverians decided to return to the completely discredited economics of the distant past. Hoovernomics would turn back the clock to ring in another Great Depression with the same old pre-Keynesian ideas that failed us in the 1930s.
As I have been arguing for years, we actually have it much better than the Keynesian-New Dealers of the 1930s had: we gave up gold and fixed exchange rates
. We adopted a sovereign currency. Our government faces no financial constraints. Except those that are self-imposed by inside-the-Beltway thinking.
*A shorter version of this was posted earlier at New Economic Perspectives from Kansas City (http://neweconomicperspectives.blogspot.com/2011/04/modern-budget-cutting-hooverians-want.html). I thank James Felkerson for producing the graph.
L. Randall Wray is a Professor of Economics, University of Missouri—Kansas City. A student of Hyman Minsky, his research focuses on monetary and fiscal policy as well as unemployment and job creation. He writes a weekly column for Benzinga every Tuesday. He also blogs at New Economic Perspectives, and is a BrainTruster at New Deal 2.0. He is a senior scholar at the Levy Economics Institute, and has been a visiting professor at the University of Rome (La Sapienza), UNAM (Mexico City), University of Paris (South), and the University of Bologna (Italy).