Why We Need a Tax AND Spending Cut

Republicans are talking a lot about certainty. But even if they had won some sort of a victory where they got the current tax rates written in stone, spending is on such an unsustainable path in terms of entitlements, it really isn’t certain at all.

That is me in the NYT. If I had had more space and more eloquence, I might have said something similar to this:

If you hate taxes, cut spending! …Short-term, uncertain duration “tax cuts” are not tax cuts at all, but deficit-financed spending.

That’s Mike Munger, economist and political scientist from Duke University. There is more here and here

What is the economic logic behind this result? Why is it that a tax cut without a concomitant spending cut might not improve the economy? There are two economic models that predict just such an outcome:

Extreme Case:

In what might be called an “extreme case”, a tax cut without a spending cut has zero effect on the economy. This is an extreme case because it requires a rather generous view of humans: it assumes we are all super-logical forward-looking processing machines (all of us, of course, except for politicians; the model assumes they don’t have a clue). The model works something like this:

Step 1. Government cuts (lump-sum) taxes without cutting spending.

Step 2.  This requires an increase in government borrowing.

Step 3.  Forward-looking taxpayers recognize that deficits are future taxes (Munger uses the helpful acronym DAFT). Because of this, they reduce current consumption in order to save for the taxes.

Step 4.  The reduction in taxpayers’ consumption completely offsets the deficit-financed government consumption. And the increase in taxpayers’ savings completely offsets government’s increase in borrowing.

In the end, switching from taxes today to taxes in the future has no effect on interest rates, national savings, current consumption, exchange rates, future domestic production, or future national income.

Economists will recognize this as the Ricardian Equivalence theorem. Non-economists will likely find this a tad implausible.

But we don’t have to rely on such an extreme model to find that a tax cut without a spending cut might not be much help. Consider another, less-extreme, model:

The Less-Extreme Case:  

Step 1.  Government cuts (lump-sum) taxes without cutting spending.

Step 2.  This requires an increase in government borrowing.

Step 3.  Only some taxpayers recognize the deficits as future taxes. As a result, these taxpayers reduce their consumption and increase their savings. But these actions only partly offset government’s deficit-financed consumption.

Step 4.  Since the public’s increased appetite for savings isn’t enough to fully finance all of the extra government borrowing, government has to get its money from somewhere. It therefore draws on two sources:

  1. Government can borrow more domestically, but has to pay a higher price in the form of a higher interest rate (under the Ricardian model, the public wants to save more, so government doesn’t have to pay a higher price). Higher interest rates make it more difficult for private investors to fund their own projects (private investment is crowded-out), lowering the nation’s capital stock.
  2. Government borrows the money from foreigners. Under this scenario, interest rates may not rise, but future national income falls because of the burden of repaying the increased borrowing from abroad.

Step 5.  Because the nation’s capital stock shrinks, future growth suffers.

Under either scenario, a reduction in lump-sum taxes—unaccompanied by a reduction in spending—fails to jump-start the economy the way politicians hope that it might.

A Big Assumption:

There is one other assumption that I have smuggled into the analysis above. Note that “Step 1” under both scenarios is a reduction in “lump sum” taxes. A lump sum tax is a tax that everyone has to pay, regardless of how much they work or consume. Economists often use it as a benchmark for efficient taxation because if the tax isn’t associated with working or consuming, then it won’t affect peoples’ decisions to work or to consume, and therefore won’t do economic harm.

It is standard for economists to assume lump sum taxation when they are talking about deficits because it makes the analysis cleaner. But, of course, taxes are not lump sum. In the real world, most of government’s revenue is derived from income taxation.

And we know from theory and data that high marginal tax rates reduce the incentive to work, save, and invest, harming economic growth. Moreover, we have reason to believe the effect can be quite large.   

So in evaluating the recently-struck tax deal, we have to weigh the “tax increases harm economic growth” evidence against the “deficits harm economic growth” evidence. In the end, I suspect we are better-off in the short-run without a major tax increase in two weeks. But in the long-run we need to cut BOTH taxes and spending. As Professor Munger puts it, the alternative is “DAFT.”