Wednesday, January 28, 2009

On this stimulus, for those who are interested

So I thought I would write a bit today because as Nick can attest, we in DC had a “snowstorm” (really just 2 inches of snow and some sleet), and no one down here knows how to drive in snow. So as a result, my office is pretty vacant today as no one came into work. So for those of you who have interest, and based on some of the convos I have recently had with some of you, I thought I would post my thoughts on this “stimulus” package but do so in a way that lays out the various arguments of for or against based on different assumptions out there. It frustrates me how the news is reporting this and how politicians are using the term “stimulus.” So here we go for those who have made it this far.

To lay it out upfront, I have been quite torn on various aspects of this as far as its economic benefit. And this is not uncommon in the economics field. Despite the political rhetoric of this being the vast “consensus,” this topic has literally been as the core of argument over academic economic debate for 70 years now, and from an intellectual perspective, this has been very interesting for me to witness in the making. In the end though, I remain skeptical.

So first, the argument in the theoretical world.

So, this stimulus package is based on a model of textbook Keynesian stimulatory fiscal policy that is at the core of any college students into Macro Econ class. So here is how it works. I’m hoping you all have a general concept of the supply and demand chart with the two crossing lines (that will make this easier). If you don’t go here. For those of you who have taken some econ or have read enough, I’m sorry to bore you on this first part.
So… the aggregate demand curve (the total demand in the entire economy) is made up of various components. The equation looks like this:

Y = C + I +G +X

Where: Y= total demand/ output of the economy (GDP)
C= consumer spending
I = Private investment spending
G= Government spending
X = net-exports

So in an economy, during a recession, C and/ or I temporarily falls (why this happens is another huge debate in economics, which I won’t go into here), resulting in lower total output Y. So according to Keynesian theory, in order to buffer the fall, government can take a deficit and either cut taxes, which will increase C and I or increase spending which will increase G. Either way, this increases total output Y (on the supply and demand graph, this shifts D to the right). Notice, the key here is not just spending as the news often seems to imply. What's important is that government takes a deficit, because if it raises taxes to pay for spending, any increase in G will be countered by decreases in C and/ or I. This is why even Obama’s Keynesian advisors are telling him not to repeal the Bush tax cuts right now. After this initial jolt that “primes the pump”, the private sector will now stop its downward spiral and resume its path or growth, where the government will then take surpluses to pay off the deficits from before. How big this chain reaction is depends on what is called the “Keyensian multiplier” which is a center of huge contention, debate and empirical work on exactly how big it is or isn’t. According to Keynesian theory, the government can continue to spend and increase output, until the economy has reached what is called full employment, at which time any new increase has to come at the expense of existing output and more spending only causes inflation. So this is essentially the basic Keynesian stimulus theory.

Now this is only true if you buy this theory. Classical/ Neoclassical economists and Austrian economists both reject this model. They say that government cannot increase output, because the government does not actually create anything, it only redistributes resources. The logic is, in order for government to spend, it must either raise taxes (which defeats the whole purpose of the stimulus even for Keynesians), print money (causing inflation) or borrow money (which the Keynesians argue for). The problem is this money isn’t just sitting there. Unless the government is borrowing money from people who otherwise would have put it under their mattress, the government is taking money out of the I component. It must take money out of the capital markets that would have gone to private investment and therefore any increase in G (or C by tax cuts), would be offset by decreased I, as interest rates rise due to government “crowding out” financial capital. So these schools say all stimulus does it cause deficits and reorder resources. A decent layman’s video (Yes, I’m talking to you Jeff), can be seen here.

Monetarists like Friedman (who's the man by the way for other reasons outside his economic work), don’t completely reject the Keynesian model, but they don’t see it like they do. As mentioned before a Keynesian thinks government can continue to spend and have a positive effect until full employment is reached. Monetarists though don’t see that as simple because it is impossible to target this spending all at just the idle resources, and therefore on the spending side, it just transfers resources that would have been used anyway already. Furthermore, any borrowing drives up interest rates and this creates a drag. The result is inflation. It was this realization in the 70's stagfaltion era that busted the Phillips Curve and the Keynesians had to respond and they altered their assumptions and became Neo-Keynesians for this issue and others. If you want to see a pretty good 5 minute video, watch here. So, because of this even Obama’s own Council of Economic Advisors estimates that the plan will spend about 7% of GDP in order to produce about 3.2% more output. Over half therefore is not new output, but just shifting output from one use to another. Therefore Monetarists generally reject using fiscal stimulus and instead focus on monetary policy via the fed (they also have different assumptions about the business cycle and see itself correcting as long the monetary policy is correct, unlike Keynesians who see the economy potentially falling into a trap less there be exogenous support). Monetarists also have a fundimentally different view of "full employment" vs "natural employment" as well as what type of monetary policy the Fed should do. If your interested read the Monetarist link.

Supply Side Economics (the actual theory, not the bastardized political version of the term), says that even tax cuts in general do not work, because what is important are the incentive forces for supplying the factors of production (labor and investment capital). So they are primarily interested in moving the supply curve (and hence the name of this quasi school), as opposed to the Keynesians who are primarily focused on the demand curve. So they want reductions in the marginal rates of taxes so that it changes behavior of workers and investors. By doing this the marginal benefit of working an addition hour of labor or investing more in a stock goes up while the cost remains the same, so more of it will happen. This concept is where the laffer curve comes from. So they would say that the types of demand style tax cuts that kenysisns prefer like mailing out a check are worthless.

There is also an issue of how strong what is called the Ricardo-Barro effect is, which is essentially, when government takes deficits, people increase their savings due to anticipating higher future taxes in order to pay off the debt. This then partially negates the increase of any stimulus, because people will increase savings as a result of it. (Barro himself, currently at Harvard whote int he WSJ recently attacking the stimulus plan),

So that’s the basic theory. Now the practical problems. Lets assume the Keynesian model is essentially correct. Well here’s where the stimulus bill goes bad even under the theoretical assumptions it was based on.

As was stated, Y can be stimulated by either tax cuts or government spending. Well first tax cuts. One of Friedman’s major contributions was finding out that people generally make spending decisions based on long term expectations of income. If you temporarily have a windfall, you will save most of it. If you temporarily have a shortfall, you will take debt and/ or tap savings. This is why the stimulus package of last spring didn’t do much. Income went up much higher then spending and people saved about 2/3 of it according to the government economic statistics. Most economists now accept this and demand side tax cuts are generally looked down upon as being effective. However, they have an advantage. They can be implemented very quickly. So we still do these, like the rebate checks of last spring and part of the new package. The savings effect is attempted to be minimized by targeting the checks at people most likely to spend them, mainly lower income or unemployed people. So things like food stamps and increased unemployment benefits are generally seen as the most bang for the buck as well and not actual middle class tax cuts.

So that leaves spending. Well the problem here is the time delay. The Congressional Budget Office (before they were strong armed by the congressional dems to take down their estimate from their website, estimated that only about $30 billion of the roughly $350 billion of spending in this bill would actually be spent in the 2009 fiscal year. This is because even “shovel ready” projects take a long time to get going. So this defeats the whole purpose. Even Keynes himself acknowledged this and said in his latter years in 1942 (his famous book was in 1936):

“Organized public works, at home and abroad, may be the right cure for a chronic tendency to a deficiency of effective demand. But they are not capable of sufficiently rapid organization (and above all cannot be reversed or undone at a later date), to be the most serviceable instrument for the prevention of the trade cycle.”

Furthermore, a more advanced look at the Keynesian model states that government stimulus should be targeted at the areas in which there are the most unemployed and idle capital. In the present case, that would be construction workers, manufacturing, mortgage finance and so on and all their supply chains. Because people just cant be retrained overnight (you can’t take a laid of construction worker and make him a doctor), if you focus the stimulus in areas that are for the most part fully employed, all you are going to do is shift resources like I had explained earlier. So that’s another major problem with this bill. The whole bill is about $900 billion. The vast majority does not target these unemployed areas. $325 billion is for education, healthcare, and clean energy (or aide to state governments to keep their programs in this area). These are all sectors with very little unemployment right now and are really the only ones that are currently growing. So any spending here will just divert an already employed person in the private sector to the government driven projects, with higher wages or course as demand goes up. About $275 billion is for demand style tax rebates which I discussed are of minimal use (unless they are made permanent and consumers begin to perceive them as such), about $69 billion is for upper middle class tax cuts via patching the alternative minimum tax again. Only about $180 billion is directly targeting areas of poorer people that will likely spend it as this is for things like food stamps and unemployment insurance. But even some of this is for Medicaid which will just transfer resources. Only about $30 billion or so is actually for new bridges, roads and so on, which is what we think this bill is about when we hear it in the news. And then there is $10 billion so in random business expensing rules with their taxes and so on. Furthermore, somewhere around $150 to $200 billion of this is not even temporary as stimulus calls for. They are new permanent spending programs. The structural deficit of the bush years was about $350 billion of so. Now we will be making it $450 to $500 billion.

There is also the issue about can the US just endlessly borrow? The Keynsisn model makes this assumption. But will foreigners continue to keep buying trillions of dollars of debt as very low interest rates?

So what does this achieve from a macro economic standpoint? It isn’t even designed that well from a Keynesian standpoint. Keynesian leaning Harvard Economist Mankiw has written a lot about this lately and its worth a read. The supply siders say this is useless; they would want say a cut in the corporate tax or capital gains tax rates. Classical Economists and Austrians reject the whole theory. Monetarists wouldn’t be happy either. So what do we have? I personally don’t completely reject the Keynesian model, but I agree with most of the practical problems with it and see it as limited. And that used to be the consensus, as we haven’t done anything like this since the late 70’s. We would just let the Fed du its thing and stop the fiscal side. There are already what is called “automatic stabilizers” in the fiscal dies of things. As people make less money, because we have a graduated income tax, people automatically get tax cuts when they fall into lower brackets. Likewise on the spending side, unemployment benefits, welfare, food stamps, Medicaid and so on all naturally go up. Empirically too, this did not work too well in the depression despite what your history teacher told you. Even Krugman, the left leaning Keynesian admits that though he liked the New Deal for its long term effects, it did NOT get us out of the depression. We did this under Nixon, Ford and Carter.. it didn’t work. Japan also did this in the nineties in their “lost decade” and it didn’t work. They just have a debt now twice the size of ours as a percent of GDP. Furthermore, we haev already had large deficits over these last few years. Would these not have been stimulative already?

In reality, this is the politicians feeling like they have to do something and they are just using this crisis as an excuse to create new spending programs that they wanted all along anyway and they just deeming them “stimulus.” Obama's own team estimates that this will "creat or save 3.5 million jobs." If you do the math, and take about a $1 trillion bill and divide by 3.5 million jobs, thats $285,000 per job! Thats not a very good deal, even if this does stimulute the eocnomy somewhar. This thing was never about stimulus like I said; this is about spending in general. Plus you know this is going to be loaded with corruption and favoritism.

We will see what happens, but I remain very skeptical about this. It stinks in theory and it stinks in application. So I hope this was interesting/ helpful for you guys. Now I actually have to go some work and I appoligize for all the typos and spelling mistakes that are pretty much guaranteed in here.

-EJB

P.S. Keep in mind that this entire discussion is about the short term effects to the business cycle; not about the long term growth of the economy.

3 comments:

JSK said...

All this goes to show is that economics is not real science. It is, at best, the freakish demon-spawn of sociology, psychology and statistics.

PS - Thank you for the "lay man" shout out. I think you used it well, because, as we all know, it simply means somebody who is not a student of the particular academic topic in which the speaker is involved. I will not feel guilty over that remark and you can all kiss my ass. Bastards.

EJB said...

Jeff,

kind of true... micro is an actual science. Macro is only a quasi science. Partially because we can't set up experiments and only have about 80 years or so of historical scenarios to test. Micro ont he other hand we littlerally have millions of examples and there is far less debate in Micro.

You can see why the "right" leaning economists are so skeptical about government interevention. Because no one can honestly say they know exactly what will happen and it is tremendous hubris to think that government can control something as vast and complicated as an economy.

EJB said...

And I was trying to get a reference to one of the Estates but couldn't get it in there. :)