(Updated and Expanded)
I’ve been thinking about Keynes lately and why it is that everyone just assumes that when government spends money, of course it helps the economy recover and creates jobs.
Let’s imagine four different scenarios.
1. Everyone in America wakes up to find they have an extra $1000 in cash in their wallet. (This money is printed by the Fed and elves put it in everyone’s wallet without anyone seeing them).
2. The government doubles the amount unemployed people receive. (The extra payments are created by printing money).
3. The government hires a bunch of workers to dig ditches and fill them back in. (The money to pay the workers and to buy the shovels comes from printing money.)
4. The government builds a bunch of bridges that are really useful. The money to pay for the cement and the workers comes from printing money.)
In scenario 1, Milton Friedman argued that people would find themselves holding excess cash. They would try to spend it. But in the short run, there is no more stuff to be had so the effect of the spending is to drive up prices. If the government persists in printing money at a faster rate than people want to hold it, some businesses may expand and hire workers but eventually, the impact of higher rates of money creation is also neutral–you get inflation but no extra stuff.
What is the difference between scenario 1 and the other scenarios? The main difference between 1 and 2 is that people who are unemployed presumably have a higher propensity to spend the money than to save it compared to the average American. Suppose they spend all of it. Every penny. No savings. The impact on the economy will again be a mix of nominal (price changes) and real effects (changes in real variables–the number of workers, the amount of cement produced and so on). Prices will rise. Will there be real effects? Doesn’t that depend on how the businesses who receive more business respond to the increase in demand for their products? Are they going to raise price or expand output? Or maybe both. That’s a reasonable guess.
The mix between nominal and real effects would depend on how much they have to pay new workers, how long they think there will be an increase in demand for their products, and so on. But confidence in the future would be a major factor. Is it unreasonable to think that what government spends the money on would affect the confidence employers and investors have in the future along with the expected level of taxes and other policy variables?
I confess that I don’t see a big difference between any of these scenarios. Nor would it seem to matter if the spending in scenarios 2-4 was financed with taxes or debt rather than printing money. It will matter down the road. Printing money will eventually lead to a higher price level or inflation. Borrowing money might do the same depending on how the Fed and the Treasury interact. And if the level of debt or the amount of borrowing is sufficiently high, borrowing could affect interest rates and further affect the level of confidence.
The usual critique of debt-financed government spending as stimulus is that it ignores crowding out by other spending–if I think my taxes are going to go up, I’ll cut back and offset some of the increased spending by the bridge’s construction workers. But the problem as I’ve argued here is more fundamental. If businesses are worried about the future (ah those animal spirits), why won’t ANY form of attempted stimulus—fiscal or monetary—be dominated by nominal effects rather than real effects?
But when there’s unused capacity—unemployed resources—won’t the quantity effects dominate the price effects? Maybe. And maybe not. Employment is down about 7.5 million since December 2007. About 2 million of those losses are in construction. For a variety of attractive and unattractive reasons, the government has systematically worked to keep the price of housing high. So there are a lot of unoccupied houses and not a lot of new houses being built. Those formerly employed construction workers have to decide whether to stay in construction or to give up and try something else. To the extent they’re still hoping the housing sector will come back, it’s unclear how any government spending is going to help them.
Whether any of the stimulus gets to those construction workers depends vitally on who gets the money and what they want to spend it on. If it goes to increase spending on health research or school teachers it’s going to mainly bid up the demand for health researchers and school teachers. It’s not going to lead to new houses. As of the latest numbers, only 5.5% of the stimulus money was being spent by the Department of Transportation. And even those funds aren’t going to create very many jobs for construction workers who once built houses if road-building doesn’t use similar people. It’s mainly going to push up the demand for road-building equipment and workers who know how to use it. Is there a lot of slack in the road-building sector?
