The Neglected Cost of Government Spending

With the news that real GDP growth in the last three months of 2012 disappeared (with -0.1% being the early estimate), calls for more government spending are again rising from Democrats in Congress. Now, they say, is the wrong time to cut spending as the economy cannot afford the additional hit from a reduction in federal spending.

Such calls for government spending and action always neglect one important facet of such policies: a cost.

Government spending may or may not help the economy; it depends on what is bought and where the money that was spent came from. However, regardless of any benefit, the costs should not be ignored. Government action always carries a cost, be it obvious or hidden.

To examine these costs, it's best to separate government spending based on the source of the money spent. Government can finance spending three ways: through taxes, deficit-financing, or by printing money. These three sources have different costs.

First, take tax-financed government spending. When the government taxes you in order to buy something or to give money to somebody else, the cost is clear. The economic benefit from whatever you would have done with the money is lost since the government took the money instead. This cost is assumedly about one dollar for a dollar of taxes raised due to lost spending and investment by the taxpayer. There are additional costs due to government inefficiency in collecting the taxes and redirecting the money into some form of spending (estimated at another $0.40 or so). These costs are not small, so unless the government does something really valuable with the money the costs of government action likely outweigh the benefits.

In the case of deficit-financed government transfer programs, the cost of the money is less clear. The person who loans the government money (through buying a government bond) is likely choosing that investment over other possible investments, not over some sort of consumption. The cost of the program is the difference in economic benefit between "investing" in a government bond and investing in actual capital investment such as building a new factory.

An investment in the productive capacity of our economy would pay benefits for years to come. When the government soaks up funds that could have been invested in the private sector, the cost is the long-run loss to our economy from lower private sector investment. Since we do not know the exact investments replaced by government debt, we cannot establish the precise cost of the deficit-financed spending. However, given typical rates of return in the private sector, the cost of this lost productive capital is likely far greater than any immediate benefit from the majority of programs on which the government spends money.

What about if the government just prints the money? This is being done today through the Federal Reserve buying government bonds. The Fed pays with money that it simply prints. This increase in the amount of dollars in our economy should mean more inflation. There hasn't been much official inflation yet (due to weak consumer demand and the drop in housing prices), although some prices have certainly been going up. Most economists agree that all this money will turn into inflation sooner or later. When it does, inflation imposes a cost on all of us by making our savings worth less (in terms of purchasing power) and in a lower standard of living unless we can get our wages to go up as fast as inflation. Since wage inflation usually lags inflation in consumer goods, printing money hits us all with a hidden cost.

Under what conditions then will things turn out better if the government spends money? The best case is when the money is spent on something permanent, say building a school, airport, or bridge. In that case, the benefits can exceed the cost. However, there is no guarantee and the cost should not be treated as if it is zero.

Economists routinely estimate the return on investment to government-financed infrastructure projects at quite high values, higher than marginal private sector investments and also higher than the economic benefit of current consumption. Thus, government infrastructure projects (roads, bridges, airports, ports) are examples of policies where the economic benefit may exceed the cost of the government policy. Some government research funding (medical, agricultural, and scientific projects, for example) also falls into this category, with benefits that may exceed costs.

However, if the government spends the money on one-time purchases such as much military spending, office furniture, government employee salaries, social security checks, and healthcare, the economic benefit is roughly equal to general consumption and likely smaller than the cost of the policy.

An examination of the federal government budget suggests that these policies are the vast majority of government spending, perhaps over 90 percent. Research spending appears to be only 4 percent of the federal budget and infrastructure is somewhere around 2-3 percent on transportation with some more hidden in other departments like education. Thus, the odds of the government doings something really beneficial with its spending are pretty low.

Hopefully, you learned two things from this economics lesson. First, politicians need to consider not just the benefit of new government spending but also the cost of that spending. Second, the probability of government spending having benefits greater than its costs is highest when the government spending is directed toward infrastructure or research investments.

Some government spending is necessary. Defense spending should occur whether its economic benefit exceeds the cost or not; the purpose of national defense is not to boost the economy but to keep the country safe from harm. However, many government programs today are being defended not on the basis of some community benefit, but as spending necessary to avoid economic damage to our economy that will supposedly occur from spending cuts. These arguments often completely avoid any discussion of the cost of the spending. These costs are real and should not be ignored.

So if Democrats push for more transfer payments and government spending that is gone as soon as it is spent, any short-term economic gains will surely be outweighed by the true long-term costs of such government policies. Only government spending that delivers economic benefits over the long-term such as research and infrastructure projects can hope to have benefits that are greater than the cost of the programs.

As we continue to debate the sequester and government spending cuts in general, remember that spending has costs. Sooner or later the bills will come due and we will realize all we have lost for some politician's short-term gain.