Orthodox Keynesians argue that in times when aggregate demand is slack, it makes no difference whether you employ people to dig holes and fill them back up or whether you actually pay people to produce public goods that are valuable. Suppose we accept this proposition – that it doesn’t matter just so long as the cash gets in the hands of the people working, then riddle me this.Why is it that these same Orthodox Keynesians argue that it is more important to have government spending stimulate aggregate demand than it is to have tax cuts stimulate aggregate demand?
For example, the economic czars in the current administration claim that their models show a 1.57 spending multiplier while only a 0.99 tax cut multiplier. In other words, each $1 of government spending magically creates $1.57 of GDP, while a $1 of “spending” via tax cuts produces only 99 cents worth of GDP. The rationale typically given for such magical numbers is that tax rebates can be saved or used to pay down debt. (ignore all of the other problems with stimulus spending for now)
My question is this, and I hope a macroeconomist out there can correct me. Imagine I am a man who is currently unemployed (because of course, government spending is always perfectly targeted at the out of work and downtrodden, it would never be spent on pensions for overpaid public sector employees or anything like that). Under plan A, the spending plan, I am hired by my town to install a bridge on a hiking trail. They pay me $20 per hour for about 500 hours of work, or $10,000 total for the project. Under plan B, rather than building the bridge, my town instead simply gives me the $10,000. Perhaps, to match the timing in A, it is paid out in smaller installments so that in terms of my income flow, I am no better or worse off in either situation.
Now what is different about A and B? The Keynesians argue that if you work for your money, you are likely to spend it all, but if you simply get a tax rebate or a tax check in the mail, you are likely to save some of it. Personally, I save more of my earned income than my unearned, but that is beside the point. Under what set of conditions can this possibly be true? Isn’t $10,000 to you $10,000 either way? And in the case of you working, while you might be forced to pay for your commute and other work related expenses, don’t you have far less time to do your spending?
My suspicion is that Keynesian economics is not only wrong on the economics, but it is just a massive accounting magic trick. You see, when government spends, say, $10,000 on me building a bridge, that $10,000 is directly added to GDP because it was a government expenditure (for those who remember, a spending identity is that GDP is the sum of consumption expenditures, investment expenditures, government expenditures and net exports). This addition of $10,000 to GDP says nothing about whether those funds were spent. If workers take all of that $10,000 of government spending and put it under their mattress, how does that get calculated? But when the government reduces taxes by the same $10,000, not all of it makes its way back into the GDP equation. In a future post, I’ll work through a numerical example of what I mean. Here is where you can start to read about it. The point is, the very construction of the GDP identity forces the multiplier on government spending to be larger than on tax cuts, even if there is no economic basis for this to be true. In fact, empirical evidence, including that by current Obama Economic Czar Christina Romer, calls this point into question.
What the heck? I get $10,000 from working, and my spending will create $15,700 of new wealth but if I get $10,000 back on my taxes, that really only generates $9,900 of new wealth. Call me puzzled.
UPDATE: After I wrote this post, this working paper came into my inbox:
Check in the Mail or More in the Paycheck: Does the Effectiveness of Fiscal Stimulus Depend on How It Is Delivered?
by Claudia R. Sahm, Matthew D. Shapiro, Joel Slemrod – #16246 (EFG ME PE)
Recent fiscal policies have aimed to stimulate household spending. In 2008, most households received one-time economic stimulus payments. In 2009, most working households received the Making Work Pay tax credit in the form of reduced withholding; other households, mainly retirees, received one-time payments. This paper quantifies the spending response to these different policies and examines whether the spending response differed according to whether the stimulus was delivered as a one-time payment or as a flow of payments in the form of reduced withholding. Based on responses from a representative sample of households in the Thomson Reuters/University of Michigan Surveys of Consumers, the paper finds that the reduction in withholding led to a substantially lower rate of spending than the one-time payments. Specifically, 25 percent of households reported that the one-time economic stimulus payment in 2008 led them to mostly increase their spending while only 13 percent reported that the extra pay from the lower withholding in 2009 led them to mostly increase their spending. The paper uses several approaches to isolate the effect of the delivery mechanism from the changing aggregate and individual conditions. Responses to a hypothetical stimulus in 2009, examination of “free responses” concerning differing responses to the policies, and regression analysis controlling for individual economic conditions and demographics all support the primary importance of the income delivery mechanism in determining the spending response to the policies.
A coarse reading of this would seem to point to support my point, no?