Are Unemployment Benefits Stimulus?

There has been much debate recently about the extension of unemployment benefits. In particular, this debate has focused on Speaker of the House Nancy Pelosi’s claim that unemployment benefits are a source of stimulus for the economy. Much of the discourse has centered on the magnitude of the disincentives to work caused by the increased benefits. However, this recent discourse is a distraction from a much more fruitful discussion about the effects of unemployment compensation on output. Rather than arguing about the magnitude of the disincentives to work, the central proposition in question concerns the determinants of consumption.

The idea that unemployment benefits might stimulate the economy is based on the Keynesian consumption function under which consumption is determined by current income. Following a negative economic shock, output and employment fall. As a result of the decline in employment and therefore current income, consumption falls as well, which further reduces output. Under such a scenario, unemployment compensation serves to increase current income and thus current consumption.

An alternative to the Keynesian consumption function is Milton Friedman’s Permanent Income Hypothesis (PIH). Under the permanent income hypothesis, individuals make consumption decisions based on their expectations of permanent income rather than current income. In other words, the future path of consumption is determined by the expectations of future income. Transitory fluctuations in income do not affect the consumption path.

Consider an unemployed worker under the PIH. The worker loses his job and his current income declines. His consumption path, however, will only be affected to the extent that he believes that the reduction in current income represents a permanent change. If the change is expected to be permanent, he revises his future consumption path downward and lowers his reservation wage. If the change is expected to be transitory, he does not change his consumption path. As time goes by, the worker continues to update his expectations.

Now consider unemployment compensation under this scenario. When the unemployed worker receives these benefits, his current income receipts increase. However, the unemployment compensation is only transitory and does not affect the worker’s permanent income and therefore does not influence the future consumption path of the worker.

The distinction between alternative theories of consumption has important implications for whether or not unemployment benefits stimulate the economy. The Keynesian consumption function, for example, supports Nancy Pelosi’s claim that increasing unemployment benefits lead to increases in consumption and output. However, the PIH casts doubt on this proposition. The question is therefore largely empirical.

A careful reading of the empirical literature suggests that the PIH performs better than the Keynesian consumption function. For example, a series of research over the last decade or so by John Seater at North Carolina State University and his various co-authors has produced a great deal of support for the PIH using microeconomic data (see here, here, and here). In fact, nearly all of the evidence against the PIH is based on macroeconomic data, which Seater finds to be the result of misspecification or problems with using aggregate data. It is also important to note that DeJuan and Seater’s Journal of Monetary Economics article finds little evidence of liquidity-contrained consumers or the so-called “rule of thumb” consumers, which are often suggested as reasons why the PIH might not hold.

Taken together with the theoretical propositions stated above, the empirical evidence would seem to cast doubt on the efficacy of unemployment benefits in providing stimulus to the economy. Advocates of further extensions to unemployment benefits can continue to argue that such extensions are beneficial by smoothing income receipts for the jobless in this lengthy recession or other similar arguments. However, it would be incorrect to say that we should expect unemployment benefits to provide stimulus to the economy.

12 responses to “Are Unemployment Benefits Stimulus?

  1. Pingback: Are Unemployment Benefits Stimulus? « The Everyday Economist

  2. The Keynesian Theory is based on a model that is incomplete. There are no allowances for where the government is suposed to get the money it uses for subsidies. Either it must print it and cause inflation or it must borrow it and cause the tax payer to need to cover the interest (and eventual repayment?).

    To support a macroeconomy where there is significant unemployment, the employees must eventually cover the expense, or as they say “There is no such thing as a Free-Lunch”.

    The claim that borrowing is a way out is based on microeconomic philosophy. The above Macro version is colder but closer to the truth.

    • Employees do eventually cover the expense, through withholding taxes levied on payrolls. Admittedly, during times of high unemployment, the benefits paid out greatly exceed the amount paid in. Of course, the converse is true in times of low unemployment, or at least may be: it depends on the amount paid into the system through the withholding tax. But that’s true with any revenue inflow/outlay situation. The point is that no one posits this as a “free lunch”, but instead as a form of social insurance.

  3. All in all, a circumspect conclusion that validates the utter uselessness of Keynesian economics…and by extention Pelosi.

  4. The amazing thing to me is that people cite Friedman’s permanent income hypothesis without reading the strange (and different ) definitions Friedman has regarding consumption, savings, etc. Friedman is speaking a different language than the rest of us -.
    For example consumption is defined in terms as “the value of services [utility] consumed during the period”. Thus Friedman’s measure of
    consumption of durable goods purchased this accounting period is the depreciation of this durable allocated to this period. The depreciation being a measure of the utility provided by the durable in this period.
    Total consumption is measured in any period by summing all nondurables produced and services puchased plus the depreciation of all pre-existing durables and depreciation of currently produced durables. [Friedman “The Theory of Permanent Income, Princeton, University Press, p. 13.]

    Savings is defined as any durable produced today that is not entirely used up today . The nondepreciated part is savings. Friedman prides himself on not defining as current consumption as including the total value of all currently produced durables such as sports cars, mink coats, yachts, etc — but only their depreciation. Friedman boasts that his taxonomy is superior to others because”much that one classified as consumption is reclassified as savings” [Friedman, p. 28].

    Under these Friedmanic definitions most transitory income received (e.g., winning the lottery, etc)will be used to purchase durables, e.g., expensive cars, jewelry (after all how many more nondurables can one purchase from the lottery winnings in the current period ?) And therefore transitory income will be mostly “saved” in Friedman’s lexicon. Moreover such savings is even better for society for it assures utility for the future [Friedman, p. 28]
    Of couse I suspect that the recipients of unemployment compensation are not going to go out and spend these unemplyment checks on the purchase of a new Lambragini auto or a gold necklace — and therefore the unemployed will not be savings out of tranitory income in Friedman’s dictionary. Instead they will likely buy food for tonight’s dinner — a consumption expenditure under Friedman’s lexicon. Hence, transitory unemploymen compensation payments are not saved and therefore the result is inconsistent with Friedman’s permananet income theory — and no empirical studies can make it otherwise.

    Obviously Friedman does not encourage people to ask the Shakespearean question “What’s in a name? That which we call a rose by any other name will smell as sweet”.
    Paul Davidson

  5. Pingback: Benefits of Whey Protein | Best Protein Powders

  6. I’m not a macroeconomist, but William Lastrapes at the University of Georgia has stated several times that there is very little evidence of a so-called Keynesian multiplier (or at least it’s a lot lower than theory predicts). Fiscal spending on things like unemployment benefits doesn’t simulate the economy the way people desperately want it to.

    In addition to the PIH, I seem to remember something called Richardian Equivalence (as MR. Chester mentions above). If the unemployment benefits are being funded by gov’t debt then we should expect to find less evidence supporting the Keynesian consumption function.

  7. Pingback: Are Unemployment Benefits Stimulus? « The Everyday Economist | The Daily Conservative

  8. Pingback: Unemployment Benefits | The Everyday Economist

  9. Rick Herrick does not know that printing money per se does not create inflation. If he readS the “Great German Inflation” by Bresconi-Torroni, the german inflation 1919-1923 was huge and so was the printing of money — but in 1924 the german government created a new currency — and then started printing money at a greater rate than 1919-1923 — with little or no inflation. Why the difference? Read my chapter on Inflation in my 2009 book entitled THE KEYNES SOLUTION


  10. What? I didn’t say anything about printing money causing inflation. I know you’re pimping your book, but it doesn’t make a great case for your book when the comment with which you’re pimping it is totally tangential to the rest of the topic. Or… are you just a spammer?

    In fact, I’m OK with stimulus spending. But my point, unrelated to deficit spending, is that unemployment insurance is just that, insurance. Insurance, by its nature, takes in more than it pays out most of the time. That surplus is, at least in an ideal system, saved and invested conservatively, allowing the interest to grow the resource pool. Then, when events occur that cause a drawdown, the system pays out. The difference between inflow, outlay, and earned interest, in a not-for-profit system, would equal zero.

    The primary difference between unemployment insurance and most other forms of insurance is that, in most insurance pools, actuarial variation is slight: there are a pretty predictable number of car wrecks and cancer patients a year and that number changes slowly if at all over time. In contrast, the unemployment rate varies widely and, when it does, the effects are across the insured pool. What this means is that car and health insurance companies have a pretty predictable and constant revenue stream. UI does not. UI also has the issue that the insured pay a lot less in premiums just as they’re collecting a lot in benefits. That makes it a perfect candidate for funding by the federal government, which, unlike most states, can deficit spend.

    Part of the problem of this analysis of UI is that it’s considered solely as stimulus measure. That’s OK, because it is of course one of the primary considerations when discussing it. But it’s not just a stimulus measure, it’s also a social safety net, which provides its own level of utility (and outside of utilitarian benefits, there’s the moral question so society supporting victims of forces beyond their control).

    So, in closing, I still don’t know what the hell you’re talking about when you say that I don’t know that “printing money per se does not create inflation,” since I made no mention of inflation and it’s possible to fund UI in excess of collected premiums in a particular year without running a deficit or “printing money,” either physically or virtually, a la QE2 (although I’d admit that it’s fairly unlikely).

  11. Looks as though the cat’s out of the bag on this… I will have a look to see whats coming….

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s