Do increases in government spending increase or decrease the number of jobs? Conventional wisdom is they will increase jobs, and a few left-wing economists, such as Paul Krugman of the New York Times, frequently are trotted out by reckless politicians and some in the news media to argue that we need more government spending in order to create jobs. If this were true, we should be able to see it in the historical evidence, so let’s look at the numbers.
Government spending grows each year, but what is relevant is whether it is increasing or decreasing as a percentage of gross domestic product (GDP) and how it relates to the percentage of the adult labor force at work. As can be seen in the accompanying chart, there is an inverse relationship between increasing the size of government and job creation. This empirical evidence, along with much other evidence, is contrary to the argument made by those calling for more government spending to create jobs. Some who argue for more government spending, such as economist Mark Zandi of Merrill Lynch, use neo-Keynesian models to justify their conclusions – conveniently ignoring the fact that such models almost always have been wrong.
What also typically is ignored by the neo-Keynesians is that there is an enormous tax extraction cost for the government to obtain each additional dollar. Estimates of this extraction cost typically run from $1.40 to well over $2.50 of lost output for each dollar the government obtains. In addition, there is vast literature showing how specific government spending programs have little or even negative benefit and, as a result, are actually wealth and job destroyers. Thus, the real deadweight loss of additional government taxing and spending is estimated to be in the $3 to $4 range.
If additional government spending could create more jobs, it would be expected that over the long run, the socialist or semisocialist economies would have full employment and the smaller-government, developed economies would have higher unemployment. Again, the empirical evidence shows just the opposite. Sweden and Canada are examples of countries that reduced government spending as a percentage of GDP 15 years ago, and as a result, both countries saw increased economic growth and employment.
The length of the periods in the chart was determined by the number of years in which the government trended relatively larger or smaller. The World War II and Korean War years were left out because of the necessary jumps in government spending as a percentage of GDP. Even during those wartime periods, there was almost no change in civilian employment as a percentage of GDP. The Vietnam War had little impact on the size of government. A big increase in government spending started during the Nixon administration after the end of the war, as many of his predecessor’s Great Society programs started to have an impact, along with Nixon’s big increase in government programs. As he famously said, “We are all Keynesians now.”
Government spending as a percentage of GDP almost tripled between 1929 and 1939 under Presidents Hoover and Roosevelt, yet the number of Americans at work fell through this period despite a growing population. The percentage of growth in government spending was less than GDP growth during the period from 1983 to 2000 (Ronald Reagan through Bill Clinton), and job growth soared.
I expect no amount of evidence will persuade Mr. Krugman and President Obama that they have it totally backward. Over the years, I have had the good fortune to know a number of the Nobel laureates in economics and have found them to be careful scholars, not allowing their political leanings to overcome what they believe to be good economics.
Mr. Krugman is the exception. He frequently misstates the good work of others. His attack on the Austrian School economists last week is a prime example. The Austrians, such as F.A. Hayek, argued that governments tend to debase currencies (when they are not tied to gold or other commodities) by creating too much money to cover excessive spending, resulting in inflation. Mr. Krugman acknowledges that there has been a very large increase in the money supply, so he asks, “Where is the inflation?” First, he greatly understates the current rate of inflation (3.4 percent over the past year), and then he ignores the fact that changes in the velocity of money (the number of times a dollar changes hands in a year) also determine the rate of inflation. Businesses and individuals are now hoarding cash because of tax and regulatory uncertainty; hence, velocity is down, even though the supply of money is up.
The Federal Reserve and the European Central Bank are creating money to bail out the indebted countries. They are either going to have to stop and destroy much of the new money they have created, or there will be a great inflation when velocity returns to normal. My bet is that at some point, inflation will come roaring back, and I also bet Mr. Krugman and others in his camp will no more admit they are wrong about that than they have about the destructive effects of increases in government spending. Some are forever blind to evidence.