As politicians have debated expansions and contractions in the unemployment insurance program, many commentators, as well as reports by the Congressional Budget Office and other government agencies, have asserted that the unemployment program increases aggregate spending, and by that channel even helps people who are not unemployed. However, that proposition has so far been the subject of few direct, empirical investigations.
Studies find that, with labor income held constant, the recipients of unemployment insurance and other transfer payments from the government tend to use those resources to consume, rather than invest or purchase liquid assets. As news articles often say, government transfers such as unemployment compensation “put money in the hands of consumers.”
That unemployed people tend to spend their benefits when they receive them is important, because it tells us that the benefits are important for maintaining their living standards. Even if unemployment benefits reduced employment as a consequence of paying people not to work, the fact that the beneficiaries’ living standards depend so much on them helps make the case that the program’s benefits outweigh their cost.
But editorial writers and commentators go too far when they jump from the consumer-spending observation to the assertion that unemployment insurance actually stimulates employment by increasing the number of people whom employers are willing to hire.
It’s true that the businesses serving unemployed people will need more employees when the unemployed spend more, but unemployment insurance is not free. The benefits are financed by taxpayers, the owners of government debt or the beneficiaries of other government programs (which cannot grow because of the size of the unemployment insurance budget).
The businesses serving those who pay for the unemployment insurance program — whether they make consumer goods or investment goods — are likely to hire less, because their customers are dedicating some resources to pay for the program.
Even if unemployment insurance did not discourage a single person from working, the net effect of the program on hiring can be positive or negative, depending on the labor intensity of the goods and services that the unemployed buy, compared with the labor intensity of the goods and services that those who pay for unemployment do not buy.
The obvious, and critical, empirical question is whether the unemployed really do spend a larger fraction of their income on labor-intensive items, and I am not aware of any empirical studies on that topic. The answer is far from obvious, For example, some of the most labor-intensive industries are hotels, coal mining and restaurants, while farming commodity crops and providing cellphone services are some of the least labor-intensive.
If the unemployment insurance program redistributes resources from people who spend a relatively large fraction of their resources on restaurants and hotels toward people who spend a large fraction on groceries and cellphone services, the redistribution may well reduce national labor demand rather than increase it.
New Keynesian models of the business cycle offer a coherent theory about how the composition of the demand for final goods might be affected by unemployment compensation that has little to do with the distinction between consumption and investment. That theory emphasizes a third spending category: the accumulation of liquid assets.
It is usually assumed that liquid assets are not produced with labor (in fact, it is assumed that they are not produced at all but rather are in fixed supply). Thus, more demand for consumption and investment goods rather than liquid assets might mean more demand for labor, especially when, as New Keynesian models assume, market prices do not adjust very well.
But that still does not tell us whether unemployment benefits increase the demand for labor, unless the unemployment program can somehow be connected with the demand for liquid assets, which empirical studies have not done yet.
An unemployment insurance program can provide some real help for the unemployed. But there is no proof that the program is good for the rest of us.