Bang for the buck or boost of the burden?

In the wake of the 2009 American fiscal stimulus package, notably little attention has been directed towards the heated ongoing academic debate on the efficiency of fiscal stimuli.

The bill, formally referred to as The American Recovery and Reinvestment Act (ARRA), consists of various tax cuts and spending measures, adding up to 787 billion dollar or the equivalent of 5% of US GDP which makes it unprecedented in size.

Browsing through popular media, it appears as if the argument for stimulus, that it creates the demand in times of recession necessary to spur employment and economic activity, represents the irrefutable truth supported by a majority of economists, rather than the highly debated theory it actually is. A good example of this type of reporting, is a 2009 New York Times article on the "new consensus" amongst economist in viewing stimulus as a "worthy step".

Taken from the NYT article, the below image shows statistical projections (grey line) as a counterfactual to actual outcomes (black line), the difference between the two represents positive effects due to the stimulus package.

Using this graph to evaluate the impacts of ARRA, without the stimulus it could have been worse, way worse.

Surprisingly, it turns out that in contrast to common belief, the economic rationale for fiscal stimulus, the so-called "Keynesian multiplier", is a strongly contested idea within academia, lacking the reported consensus support.

A pretty poignant expression of this divide in the support of the fiscal stimulus, was expressed in a New York Times/WSJ ad. The ad, sponsored by freemarket oriented "Cato Institute" contains the names of 200 highly acknowledged researchers, including Nobel Laureates, all petitioning against the stimulus package.

The rationale for stimulus offered by economists is often expressed using the Keynesian multiplier effect of government spending during recession.

Essentially, the"multiplier effect" can be explained as follows: in times of recession there is a lack in aggregate (summed up) demand due to "pessimism" (i. e disbelief in the economy) . This results in so-called idle resources, that is resources (workers capital...) which could be working but are not due to recession based lack in demand.

Thus the economy is in need of a boost in demand to activate these idle resources. This is where the government steps in and creates demand through investing in various projects, such as improving infrastructure by building new roads or, as in the case of Germany, renovating Humboldt University.

In one word , the government stimulates demand through interacting and investing in the economy. The presumed positive effects of fiscal stimulus does not stop with putting idle resources such as workers back to work. The multiplier effect kicks in as these newly employed workers start receiving wages and start consuming and demanding even more goods.

Thus in net effect, if the Keynesian multiplier holds, one dollar spent by the government in fiscal stimulus could end up being 2 or 3 or even 5 dollars of effective economic activity, the multiplier thus refers to how much you have to multiply the original gov spent dollar value with to get the final value of this fiscal stimulus. For example, a multiplier of 1.5 would indicate that a dollar spent by the government leads to 1.5 dollar worth of economic activity.

The counterargument to this theory can be expressed by the "Ricardian equivalence" theorem, which states that it is impossible for the government to spur economic activity through debt financed fiscal stimulus, since the consumers internalize the gov debt into their own budget. Therefore, consumer choices determined by a utility maximization process, will not change as the budgets that this maximization is subject to does not change.

Intuitively, if I consider the gov debt my own debt, a 300 debt financed tax rebate, or a (debt financed) fiscal stimulus paycheck will not make me demand more goods, as I know that I have to pay this extra money back in the future, either through higher taxes or inflation. According to these theorist then, debt financed fiscal stimulus will actually make me consume and demand less today as I will try to save up for paying back my debt in the future.

Thus if the Ricardian equivalence holds, a debt financed stimulus package will have the exactopposite effect than the intended and will actually lower economic activity, in terms of "multiplier language" then, we would have a Keynesian multiplier of less than one as stimulus would crowd out private investment.

It might seem as if the assumption that consumers think this far in their decision making process it pretty unrealistic, however several studies have shown that a temporary change in disposable income does not change consumption behavior, as individuals base their spending on permanent income rather than temporary pecuniary windfalls.

For instance this graph plots DPI, disposable personal income against personal consumption, clearly consumption did not rise as a response to the ARRA financed positive shock in income.

Another noteworthy contribution to the stimulussceptic literature, is Robert Barros "Macroeconomics effects of government purchases and taxes" in which Barro, using WW11 data ,estimates that the peacetime multiplier is essentially zero.

Returning to the image at the top of the this post, consider this extension created by John B Taylor (yes the man behind the Taylor rule):

As you can tell , this extension, with projections based on New-Keynesian models and the Barro paper, shows an impact of ARRA insignificantly different from zero.

One fairly vivid and entertaining example of how disparate the outlook for fiscal stimulus is, can be seen in this UC Davies debate in which a dataladen, full-of-proof contra- Prof Boldrin is debating with fairly unequipped and mostly-metaphorical pro-Prof DeLong.

Banging or burdening, the debate on fiscal stimulus in my opinion is an extremely interesting, challenging and thought-provoking one. Surely there are valid arguments for both sides, with this brief presentation, I am hoping to open up for discussion.

Also on this note, I think it is important to note that the above analysis only deals with the supposed economic benefit of stimulus, if we were to look at social benefits it is obvious that extension of various social security programs are going to increase the welfare of its recipients.

However, as politicians use the economic rather than the social argument to support stimulus, the potential flaw of such an argument is a highly salient topic for future policy analysis.

Ok I am going to end this post for now and start at a new one discussing what can be seen as a proxy for the arguments of the prostimulus side: the Council of Economics advisers quarterly report on the impact of ARRA. For interested readers my next post will specifically be on the statistical baseline, presented in the report and used as a benchmark for ARRA stimulated growth.

Tschu