Oh yeah, that too

In the last post, I concluded that the macro debate today is about whether or not at the zero lower bound monetary policy can be effective. Theory says yes; data is inconclusive.

I neglected to mention that the debate is also about whether or not fiscal policy is effective at that lower bound. Martin Feldstein, in a highly readable and short paper that anticipates and takes the reader through this whole debate, says fiscal policy can be effective in such a situation (search google scholar for a free version).

Given we haven’t seen deflation so far, fiscal stimulus is a hedge against it. Once stimulus finally comes on line, if the economy is in a deflationary spiral, it will help get it out. Of course, we all trust that if that contingency doesn’t come to pass, the spending will be quickly unwound. Right? Right!?

20 thoughts on “Oh yeah, that too”

  1. Well isn’t that the Great Depression debt deflation story of Bernanke?

    In any case, I’m not saying such a thing will happen… I’m just trying to make sense of policy.

  2. anyway, don’t worry, once we deduct what’s going on at the state and local level, the “giant stimulus” comes to a whole 1% of gdp…

  3. What I don’t like re: the world “spiral” is that it comes from a ad-hoc dynamic tacked on crappy old models. Debt deflation is about wealth effects, which I can understand.

  4. Mr. Veblen, I’m not worried about anything. I want to understand the economic rational for fiscal stimulus. The profession is unanimous that fiscal stimulus is inferior to monetary policy in normal times, when the interest isn’t zero. I’d like to hear a story that says this policy preference should be switched when we’re in a liquidity trap.

    Also, I’d like to hear a theory for why “infrastructure spending a year or two from now” policy is better than monetary policy in a liquidity trap. If you say that it increases inflation expectations because of concerns the Fed will monetize debt, you have to give up on the “stimulus” aspect of such policy and all the hullabaloo about multipliers was wasted breath/typing. Otherwise, I’d really like to hear a story that suggest “infrastructure spending a year or two from now” increases aggregate demand today.

  5. OK, first I agree with you that the timing of the stimulus has been completely mishandled, and shows real incompetence among Obama’s top economic advisors, specifically Larry Summers and Christina Romer. Why do I say this? Because they scheduled, on purpose!, some of the funds for states and some of the tax cuts to come in 2010. Whether the actual stimulus money could not have been spent faster or not, this much is clear: because states and local governments have massive shortfalls, even after all the stimulus money, they are cutting back hard on money right now… For example, Pennsylvania will not pay state employees for the month of July until later… California has paid out 150,000 “IOUs”… Of course, the Federal Stimulus did greatly reduce state budget shortfalls, and as a result, California was only forced to raise taxes by about $14 billion and decrease spending by $15 billion for the period just started thanks to the stimulus money, but it was not enough to avoid paying with IOUs instead of real money. In retrospect, however, it is quite hard to believe anyone was ever opposed to stimulus to states in the first place! But this is precisely what the moderates in the Senate cut…

    The economic theory goes that when people have less money in their pockets — their disposable income suddenly falls — they will spend less. Now, yes, I completely agree that Bernanke could compensate this decline in disposable income with quantitative easing, but it’s a rather weak measure b/c it all just goes right in to banks’ excess reserves. The money multiplier dropped below 1 for a time! Conversely, when the Treasury floats more T-bills, it will eventually just come back out of banks’ excess reserves…And even though Bernanke has done a lot of this (the Fed doubled its balance sheet overnight!), it was not enough to keep unemployment from skyrocketing toward an expected 10.5% at year end. In other words, the QE done by the Fed failed to prevent this (deep!) recession. I think he screwed up, and should have done/be doing much more, but I have not heard any mainstream economists go on record voicing that opinion (nor you!). So, QE seems to have much less direct effect on the economy than a typical interest rate cut. (And, to the extent you do not believe this, then you must believe the Fed has been willfully negligent… to allow a severe recession even when its policy tools were perfectly effective.)

    Though I would have planned a stimulus consisting of, say, $250 billion in immediate stimulus checks, $250 billion in immediate transfers to states and cities, and another $250 billion in longer-term investment projects, the longer-term investment projects are still defensible — next year we still expect unemployment to be high. Intrade now says unemployment will still be above 10% for all of next year. The Fed will likely keep interest rates pinned to 0 over the same amount of time. So, even though I would have liked much of the spending to come sooner, anything spent in 2010 will still be useful.

    But opposition to the original stimulus implies advocating deep cuts in government spending and increases in taxes in the middle of a recession. This is the position that you, and many other moderates and conservatives, took, and the one i have not seen defended. Indeed, stimulus advocates (including this blog) have seemed utterly unawares of the fiscal position of the states. (I remember your post where you said the stimulus came to 5% of GDP, when the net stimulus, minus what is going on at the state and local level, is far, far less…)

    And, dammit, for the last time, in general, large deficits are inflationary. This means the Fed will generally respond, soon if not immediately, by raising interest rates, and long-term bond yields will also increase due to expected inflation. This will kill off the stimulative effects of large deficits. Now, the whole goal is to generate inflation!

    Lastly, some 80% of the stimulus will have been spent in the next 18 months (after which time we’re still expected to be at 10%+ unemployment!) When we start to talk about the stuff that will happen “two years down the road” and divide it by GDP, we are really talking peanuts…

  6. Well, my reading of the CBO report is that some $628 billion out of $788 billion will get spent by then end of next year. (Next year year, not next FY.) Some 7/8ths gets spent sooner than two years from today.

    I’m not about ready to concede that there will be any net stimulus on behalf of the government more than two years out, however, as in the past few recessions, states’ budgets have taken a long time to recover, just like the labor market. States will likely be pulling back thru’ 2012…

  7. I agree with Thorstein that the fiscal stimulus should have been spent sooner. Some of this was probably to ensure that the funds get spent right, “haste makes waste.” But still this is not an excuse for the long waits for most of the spending.

    As for the delay, unemployment will almost surely be still very high at that point. We still may not have reached peak unemployment. What level of unemployment would make it not worthwhile for the fiscal stimulus to take place? 8% 7% 6%?

    Granted, I’m starting from different assumptions than you are. I don’t see fiscal stimulus as just inflationary. It can create jobs. That’s why lots of senators didn’t want to can the F-22 program, because it will cost jobs. (It still should be canned, but that’s a separate issue)

    When the government hires someone to repair a road who was formerly unemployed, does that lower the unemployment rate or not? This will not necessarily be inflationary. If not, then should unemployment be lowered? And if so, how?

  8. Yes, of course it should have been spent sooner. We get the fiscal policy our institutions will implement, not the ones we want. Our particular institutional framework (i.e. congressional budget committees) favors infrastructure projects over aid to states. Infrastructure takes longer to implement.

    There’s a shallow slope on the line fitted between unemployment rates and funds from ARRA (here’s the SF Fed analysis). The fiscal policy our institutions implement doesn’t do a good job targeting idle resources. It would be better if it did, but it doesn’t.

    I hate to be a dick, but those that supported the ARRA are the baptists and the special interests in congress the bootleggers in this process. There should be more dissertations written on political macroeconomics.

  9. I don’t want to be a dick either, so I won’t make much of the fact that you contradict the article you cite:

    “So, while ARRA’s state allocations do not represent the absolute optimal stimulus, they are on the whole well directed. Overall, that means that the economic impact of this support for state governments is more likely to exceed than to fall short of forecasts.”

    The question becomes, why did you reach a different conclusion than the SF Fed?

    Infrastructure is about $111 billion in the recovery package. You seem to support aid to states in principle, which is 144 billion. Do you support the tax cuts too? Because then you would support over half of the ARRA. My point is that I see a lot of people opposing ARRA because they oppose government spending, when the majority of the package is tax cuts and aid to states, which many people support.

    If you don’t mind, what would your optimal ARRA look like (or should we have a fiscal stimulus at all)? For me, a good start would be halving the tax cuts and increasing the aid to states to 500 billion. If more good infrastructure projects are available, those could get more funding too. I would rather they get done in time, but given the terrible shape of American infrastructure, they couldn’t hurt at any time. http://www.infrastructurereportcard.org/

  10. The Fed analysis says the correlation is negative between unemployment and ARRA funds…

    Optimal fiscal policy, ignoring implementation constraints like the Congress and timing, would be immediate and targeted at idle resources. Our institutional framework does not produce policy like this… implementation constraints are binding.

    Optimal Will Ambrosini transport technology, ignoring implementation constraints like physics, would allow my instantaneous transport to and from any position on the globe. Unfortunately, those constraints bind, too.

  11. Eh… no? You can change institutions but you can’t change physics…

    Costs contingent on particular institutional arrangements are not proper social costs. Otherwise, the monopoly unable to perfectly discriminate would be efficient, for example.

  12. Um, the big “institution” in this case was the conservative economic ideology of Christina Romer and Larry Summers who chose to start small with direct transfers to states and include corporate tax cuts, and the “centrists” in Congress who cut aid to states over irrational concerns about the deficit.

    I’m not sure they would have been disabused of these concerns had they been reading this blog…

    Had liberals been in charge, California and Pennsylvania would not be paying IOUs right now…

  13. Gabriel, that’s my point. You have to change the institutional constraints, not bitch about the same bad policies that the same bad institutions produce.

    Thorstein, you can always say “if only we had the right people in office…” Our system puts these people in office and they come up with these sub-optimal policies. And you’re crazy to think those two people were responsible for the shape of ARRA spending. Krugman or Stiglitz’s force of will would have overcome the demands of the pork swilling Congressional committees? Their crusading would produce the necessary political compromise, votes and a President’s signature?

  14. Let’s one-up this biatch!

    Current constraints are the result of the current political-economic equilibrium. No one can magically “change” them unilaterally since, by the definition of equilibrium, you’d have to be worse off.

    Calls about changing institutions, then, are as misguided and futile as calls for better policy in the abstract. [I’m not sure I believe that but I see how that might be true.]

    So, what one needs to do is pray [if one is into that] for some *exogenous* change in fundamentals that would translate into a positive change in institutions, which would translate into a positive change in policies and, finally, outcomes.

  15. The correlation is negative between unemployment and *infrastructure* funds. Big difference.

    And the report states:

    “However, all states have seen substantial increases in their unemployment rates since the start of the recession, so it’s hard to argue that any state currently is at full employment and does not have idle capacity that can be put to use on major construction projects.”

    You don’t have to state what your optimal fiscal stimuls would be, but it’s easy to say that government is highly imperfect. It doesn’t add much to the debate though. Who si going to disagree with you? It’s the blog equivalent of a cheap shot.

    But if the alternative is Austrian style inaction and liquidationism, then I’ll take the ARRA any day.

  16. Piero, my argument is over the intensive margin not the extensive margin. Yes, there was targeting, but it was very poorly targeted. Also, I’ve stated a number of times my version of optimal policy, but any discussion of optimal policy has to be conditioned on all relevant constraints to the planner’s problem. If there were no political constraints, optimal policy would be implemented quickly and targeted at idle resources. Additionally if you believe Christiano, Eichenbaum, and Rebelo and the typical assumptions New Keynesians make, then fiscal policy only makes sense when the spending is done while the economy is in a liquidity trap.

    Call it my subjective opinion if you must, but the ARRA was neither quick nor was it well-targeted. I base my subjective opinion on the CBO analysis. Further, given most spending is next year or beyond and by my reckoning there’s a slim chance we’ll be in a liquidity trap next year then the ARRA won’t even pay for itself. If the ARRA represents the best fiscal policy our institutions can provide, this suggests our political institutions are incapable of producing optimal policy or policy even close to ideal.

    Gabriel, the exogenous shock could be this new knowledge about our current set of institutions. We now know they’re incapable of producing good fiscal policy. We didn’t know that (the macro student in me says, “yes we did!” but keeping my sanity in these debates requires me to ignore my inner macro student). So now we can use this new information to find a new equilibrium.

  17. I think I understand better now your position. I meant composition of the ARRA and you seem to mean implementation, but that’s fine.

    Regarding the liquidity trap, in theory a zero interest rate policy can be optimal in a Friedman optimal deflation model. Many people seem to equate this with a liquidity trap, though I don’t know that they necessarily need to be. If money is neutral, the zero interest rates is definitely possible with deflation and positive real interest rates. A liquidity trap, which I think is not very well defined, for me is when monetary policy is ineffective, as excess reserves make the money multiplier plummet and blunt the effect of monetary policy.

    While that seems to correspond to times when the economy is performing poorly and when interest rates are low, I don’t see how a Zero interest rate and a liquidity trap are equivalent. With no deflation and no interest paid on reserves, there would not necessarily be a liquidity trap with zero interest rates.

Comments are closed.