Skip to content

What the Future May Bring

February 8, 2009

The French Cowboy was wondering why Obama would publish this analysis of the possible effects of his economic “stimulus” on the job market. It goes so far as to make projections on the number of jobs that might get created in specific industries and even how many jobs might go to particular demographic groups. Ironically, the parameters used to calculate the latter are less questionable than those needed to calculate the overall effect on the job market. Forecasting the total number of jobs that result from the programme is the tricky part. Once that number is established it is comparatively easy to estimate how those jobs will be distributed.

The sentence that struck the French Cowboy as especially weird was this one from page seven: “Our analysis assumes that households treat the tax cut as permanent in determining their short-run spending.” They assume that families will react to short-term tax cuts as if they were lasting tax cuts? Typically (and easy to understand intuitively), a short-term tax relief results in less spending than a lasting tax cut. Romer and Bernstein even state this themselves on the same page. This means that their estimated short-term effect on (spending and therefore ultimately) jobs through the short-term tax cuts is too high. Why didn’t they correct this?

Another odd thing is burried in footnote 2. Romer and Bernstein assume that the job effect of an increase in productivity is roughly the same no matter how productivity gets sparked. Fair enough. (If this is true there are other, non-economic factors to look at that would help to evaluate the different methods.) But then they add this:

[S]imulations using private-sector forecasting models suggest that the number of jobs created by a given increase in GDP is likely to be slightly higher for broad-based tax cuts than for infrastructure spending, presumably because the jobs created by infrastructure spending pay higher wages on average. This effect is small, however, and does not reverse the conclusion that a dollar of infrastructure spending is more effective in creating jobs than a dollar of tax cuts.

The French Cowboy could have accepted it if they had said “the effect is small, so we just ignore it”, yet they chose to say “the effect is small, yet the conclusion stays the same.” There is a tiny difference between the two. They write the latter but they mean the former, because the latter doesn’t make any sense. Romer and Bernstein claim that as a “starting point” they assume all increases in GDP to have the same impact on job creation (see footnote 2). Yet they also state that “[t]he direct spending programs have the largest job bang for the buck.” (page 7) and that private sector models showing that tax cuts have a stronger job-creating effect than infrastructure building doesn’t “reverse” their opinion that, in fact, the contrary is true.

Of course, this assumption is at the core of the debate about what the stimulus bill should look like: government spending versus tax cuts. Conservatives happen to believe that ceteris paribus (ie “all other things equal”) allowing households to spend their money the way they want to beats government taking the money and spending it on whatever they want. It is a matter of freedom and individual choice and about the sad truth that governments tend to be horribly inefficient and somtimes even corrupt. Romer and Bernstein have used an arguably discretionary leeway in favour of government spending and unfavourably towards private spending (or tax cuts) even while they cite research results that point into the opposite direction.

But there is a more general question about this piece. Even if a programme of the exact mold as the one Bernstein and Romer have used for their analysis were to be implemented it is highly unlikely that the actual effect would turn out to be even close to what this paper projects. The underlying parameters are simply too uncertain. Bernstein and Romer know this. Obama probably knows it. So why do they put these hard numbers out when they will (most likely) be used as a benchmark for the actual outcomes of M Obama’s plan? It doesn’t seem to be a smart move, especially since it appears that this analysis tries to make the effect of an Obama economic stimulus look better than it is likely to be.

But it could be a strategic step to put out these estimates. President Obama might use any discrepancy between his team’s forecast and the actual results of his policies as an argument for future interventions into the economy. And he will most likely find culprits for any shortcomings outside of his staff and party to keep the political price for himself low. The Republicans’ tinkering with his original proposal, for instance, or their insitance of discussing it and thereby delaying it. So Obama might have created an excuse to meddle with the private sector once the effects of the “stimulus” fail to show (this is purely hypothetical, of course) and since Romer and Bernstein’s analysis is based on a plan that will be altered through House and Senate amendments before it gets passed, Obama can always distance himself from the estimated numbers.

No comments yet

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 )

Google+ photo

You are commenting using your Google+ 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

%d bloggers like this: