When the press and the politicians talk about the effects of the stimulus package, they usually operate under a misunderstanding of the post- Keynesian economics that supposedly underlie it. Most argue that if, say, the bursting of the derivatives bubble wipes out a trillion dollars from the economy (and it was actually more), the government has to step in temporarily with a trillion in government money to take the place of the lost money until investors, buyers, sellers, markets etc. can start buying and selling again. The idea that is wrong is that a trillion needs to be spent because of a trillion was lost. It’s not exactly true because the two monies are not exact equivalents. If the government puts in a trillion, it will not actually lose one trillion and it stands to make back a portion of what it does spend. This is for three reasons.
The first reason, let’s call the “flushing effect.” Say I get a job from the stimulus and I spend a hundred dollars I wouldn’t have otherwise spent on groceries. My local store makes that money, so it uses it to order more veggies and canned goods from its suppliers, who make some money, and they, then order more stuff from their processer who makes some money and buys raw veggies from the farmer conglomerate, who makes some money and orders more seed, fertilizer, gas, and doesn’t fire his farm workers, who then get to keep their jobs and buy more groceries. And so on. The money flushes in and around the system and everyone makes a little money off of my original one hundred bucks.
Incidentally, this dynamic is one of the arguments often used against the maquiladoras just over the border in Mexico: the workers make a little money (not to be sneezed at, as my illustration shows), but by far most of the money, supplies, raw materials, and profits all stay in the US. So the “flushing effect,” is small.
A variation on the flushing effect is sometimes called “Dynamic Scoring.” It says that the government is planning on spending a lot of that trillion on hiring people for those “shovel-ready” jobs who presumably were previously out of work. Since, roughly speaking, about forty to forty-five percent of the people who are out of work also collect unemployment, therefore the trillion in expenditures immediately drops by about twenty-five percent to $750 billion because these guys (mainly guys) are no longer collecting that unemployment. Plus, since they would now be paying taxes, the stimulus expenditure drops another fifteen percent (more or less) to bout $600 billion. Then, add in all of the money that is flowing through the system caused by the fact that they now have more money in their pockets and the flushing effect takes over, most of which will pass through a sales tax, making even more money for the government. So, for that one trillion, the government actually lost about $750 billion and then made back (in income tax and sales tax) another $100 billion. These numbers are rough, but few economists would argue with the principle behind them.
There is yet another variation on all of this which is sometimes called the “Multiplier Effect,” and often attributed to John Maynard Keynes (though it never showed up in any of his writings). Paul Krugman, who has a clever way of words, calls it the “more bang for the buck” effect.
It’s a bit complicated, so read close. Let’s say you wanted to put a bucket of money into the economy quickly (as in the “Economic Stimulus Package”) and you wanted to get the most “Bang for your Buck.” Here’s the textbook economic formula for doing that: 1/(1 - c(1-t)). That’s cool, but what does it mean? Start with the “1” that’s the input into the economy from the stimulus. Let’s say that it’s a trillion dollars. The “c” means “consumption,” the marginal propensity of people to consume out of what they have received in income. The “t” is taxes, your marginal tax rate. So, now let’s say the “c” (consumption) is 0.5, you consume half of what you receive in income, and the “t” (tax rate) is 1/3, high for some, low for others. Then if you divide $1 trillion by the amount of one trillion minus the 0.5 consumption of income, and your (aggregate) income is the one trillion less the 1/3 tax rate, then you come up with a “bang for the buck” multiplier of about $1.5 trillion. Is that clear? What it means is that for every one trillion that the Obama administration (or any administration) chucks into the economy, because of the way we consume and tax, it will flush around and create $1.5 trillion in value, in stimulus. So, assuming that other variables don’t get into the mix and mess up the formula, you should raise the GDP by roughly half of what you put out in stimulus money. And remember, a lot of that $500 billion would come back to the federal government in taxes later. So, not only are you not spending as much as the pundits said you were, you are also getting almost half of it back again in taxes. Not a bad deal.
So, now that we understand how the stimulus should work, what would be the best way to make it work? What is the best way to get the money flushing around in the system?
There are basically three different ways to get money into an economy. The Federal Reserve does it by lowering interest rates (which are now effectively zero, so forget that for a while). The Republicans tend to do it with tax cuts. And the Democrats tend to do it by direct money to things like jobs or tax incentives. Which method gives the most bang for the buck? Liberals and conservatives are a bit divided on that. All (well, almost all) of the standard textbooks would suggest that Keynesian direct public spending works best. However, there have been studies (a few studies) that seem to show that tax cuts work better. I’ve looked closely at one of those studies and it did clearly find that a slight bit more money “flushes” though the economy with tax cuts than with direct inputs, so to claim a better buck banging for tax cuts is not just partisan “smoke blowing” by ideologues. There is actually real evidence for something akin to Republican dogma on tax cuts.
However, here are two caveats. First, the tax cut that creates the most bang is to the lower and lower middle income brackets. When a rich person gets a tax cut it has to be a whopper before it will have much of an impact on his or her lifestyle. But when a poor person receives a tax cut (or a financial reward of any kind) he or she spends it on milk and eggs (or whatever). (In fact, unfortunately, some poor people who do not balance their budget well tend to spend more than their tax cut and have to reign in their spending later on, but that’s a different problem). And the higher up the income scale, the more the tax cut is saved rather than spent. One could argue that savings are good, but not when the money is needed for stimulus. Note how the tax rebates of the spring of 2008 pretty much did zip when enacted. Nobody but the very poor spent the rebate. The rest of us more or less stored it in the bank for a rainy day (which has, by the way, been happening since about September).
The second caveat is that tax cuts that work best for the vast majority of us are targeted cuts that give us incentives to buy things. Cuts for college tuition, for example, or for small businesses wanting to expand for another. Generalized tax cuts that generally (though not always) tilt toward the upper income brackets tend to be the worst for stimulus. That’s why a lot of people were disappointed in the Obama administration for allowing over a third of the stimulus package to be tied up in general income tax relief.
Oh well. The bang for the buck portions in the stimulus package are good.