Reihan Salam has a quick piece in the Daily on some problems with the new stimulus bill.

I want to talk about each one. Let’s go

At the heart of the president’s proposed American Jobs Act is a $240 billion extension and increase in the size of the current payroll tax cut. . . .

There is, however, a great deal of empirical evidence suggesting that workers save the proceeds from temporary tax breaks, particularly people who aren’t optimistic about their future economic prospects.

I tend to think the empirical work here over pushes the point. Some of it for example asks people whether they saved or spent their tax rebate checks. Well, of course almost everyone saved it. It would be impetuous to do otherwise.

However, the real question is whether their net spending was higher or lower due to the tax rebate. Were they more free to spend other monies or did they feel ok about dipping into saving to make a down payment on a car given that their savings buffer was now higher. That’s what you are really looking for.

We wouldn’t necessarily expect people to run to Best Buy, check-in-hand, and get a new TV. Though a surprising amount of people appeared to do just that.

Moreover, in our particular crisis which is in large part a balance sheet crisis, increasing household savings will also help to ease the recession. It won’t do much in the short term but it should make the recession end sooner as people repair their balance sheets sooner.

And remember that the payroll tax pays for Social Security. Money diverted from that program has to be made up with higher taxes in the future. As America ages, those higher taxes will be imposed on fewer workers to support more retirees. The math gets dicey very quickly.

The math only gets dicey if people insist on continuing this accounting gymnastics known as the Social Security trust fund. I prefer to think in real resource flows. The government takes command of resources through taxes, it releases command of resources through Social Security checks. All of that happens in the current period and should be analyzed in current period terms.

However, if people want to stick with the gymnastics then you simply have to add more gymnastics to make the balance sheet balance. The Treasury will pay the trust fund on behalf of the public. Or we could get really fancy and the Treasury could pass a wage-subsidy equal to half of your payroll tax. It could then allow that subsidy to be transferable in lieu of taxation. So, workers get a subsidy which they use to turn around and offset their payroll tax bill.

No change to the trust fund.

More generally, however, any effort to cut taxes now will result in higher taxes in the future as long as the government doesn’t change in size. However

  1. Tax cuts have macro effects only when we are up against the zero lower bound. Otherwise the Fed can always offset the contractionary effect of a tax increase.  So lowering taxes now and raising them later can make GDP higher all the time
  2. The real resource constraint decreases over time. Fundamentally this is because the federal government runs a positive spread on its borrowing. Or in terms that I like to think in, right now the carry on taxation is less than the yield. I know that point can be a bit much for people to swallow. However, it should be clear that when real interest rates are negative, you wind up paying back less than you borrow.

Moving on

The president called for new spending on education and infrastructure. There is an excellent case for infrastructure spending, particularly if it’s part of a long-term plan that includes finding smart ways to pay for it. But as Alice Rivlin, President Clinton’s OMB director, has said on numerous occasions, infrastructure spending is not the best way to stimulate the economy in the short-term. President Obama seemed to acknowledge that fact last year when he said, “There’s no such thing as shovel-ready projects.” Somehow amnesia has taken hold.

I more or less agree with the thrust of this but I think the conclusion is odd. We should have infrastructure. Borrowing costs are cheap. But, we should consider it a bad idea because its not good immediate stimulus.


I mean yeah, if the choice is between infrastructure or a bigger payroll tax cut then I choose a bigger payroll tax cut. However, if I have gotten as much payroll tax cut as I can get and someone is also willing to give me infrastructure then I will go for it.

Especially since it looks like infrastructure in the US is aging and thus replacing it will yield a positive return, where real borrowing costs are again negative.

The stimulus effect might come late, but we might still need it then and besides we get better roads and cheap financing. That seems on net good not bad.

Now to Schools

Given the amount of money that is wasted in the education sector, there is little reason to believe that a new federal windfall for public schools would be spent wisely. Cash-strapped school districts have finally started to cut spending wasted on bloated administrative budgets, lighting empty classrooms and much else — money that belongs in the classroom, yet somehow never makes it there.

This is a more nuanced public choice problem. Here the choice Reihan lays out is between short term macro stimulus and long term reform. And, I think he is probably right.

The issue is whether the costs are worth benefits. This would require more serious analysis. However, we should keep in mind that it is unlikely that reducing resources for schools – which is the effect of a recession in a balanced budget environment – will result in pure efficiency gain.

Its likely that the same forces which made schools inefficient in the first place will continue to assert themselves even during the downturn and only starve classrooms more.

Now you may get more pushback from local official and crisis may spur people to action. So, this effect has to be counted as well. My end take on this issue is that it’s hard to tell.

On tax loopholes for oil companies

The president did say that he wants to reform the corporate tax code, which is great news. But what kind of reform does he have in mind? Incredibly, the president used a jobs speech to make the case against “tax loopholes for oil companies.” To translate this into language we can all understand, the president is calling for increased taxes on drilling new oil and gas wells

I tend to agree with Reihan here, though as a matter of principle we shouldn’t be attempting to push the economy towards any particular energy sources. However, as efficiency clean-up measures go, getting rid of tax credits for domestic oil drilling would be, to paraphrase Lewis Black, on page six right next to “are we eating too much garlic as a people.”

Moving on to housing

To aid homeowners in distress, the president shared his intention to help refinance mortgages at today’s low interest rates in order to give families as much as $2,000 in extra spending money. Alas, there still is no such thing as a free lunch. As Harvard economist Edward Glaeser has explained, this refinancing effort would lose investors in the mortgage market a large amount of money in the short term and it would pay dividends to homeowners over the decades-long life of a mortgage. Because many mortgage investors live in the United States and might otherwise spend that money they’d lose in the process, this measure might actually destimulate the economy in the short term.

I am going to go ahead and say I haven’t read Ed’s piece yet, so maybe he makes some good non-obvious points. However, the basic mechanism outlined here seems “silly” to me. Silly is not quite the right word but I am not sure what is.

So, whatever gain homeowners receive is roughly equivalent to the loss that bondholders receive. If the bondholder wants to count all loses as coming in the current period he can. If the homeowner wants to count all gains coming the current period he can. But, in both cases it’s the coupon that has changed.

Now there are two potentially important effects.

1) Being able to ride underwater homeowners is a positive of buying a Fannie/Freddie bond. You don’t have to worry about default because the bond is insured. However, if the homeowner is locked out of refinancing then the homeowner looses a valuable refinancing option. That’s good for you as a bondholder. Knowing that, you as a bondholder are willing to accept a slightly lower yield on the bond. If the government signals that it is willing to take that advantage away it will raise the cost of borrowing for future households.

2) Lowering the current cost of mortgages has an externality effect. It means that fewer homeowners will default, which means there is less distressed inventory coming on the market, which means there is less pressure on home prices, which means fewer homeowners will default.  So just as a wave of defaults can build on itself, a wave of non-defaults can build on itself.

Those two effects cut the otherway.

However, for the stimulative effect you are causing losses to folks who are by and large not liquidity constrained in exchange for gains to folks who are almost by definition liquidity constrained. This will almost certainly then be stimulative in a short run macro way.

So on net I would say that once again this isn’t the stimulus that I would have designed but it is much, much closer to what I would have designed and I think its strongly workable.