Making money at zero cost

Last fall, I became convinced we need to get US nominal GDP back up to its trend level. Nothing has changed since then, except that my impatience with the Fed has overcome my blogging energy barrier.

Let's do a helicopter-drop style thought experiment. Congress raises the debt limit. The US Treasury sells the Federal Reserve Bank $3.1T in special, 999-year bonds. To "pay" for the bonds, the Federal Reserve Bank simply types "new account... balance: 3,100,000,000,000.00" into a computer. As always, the Fed remits all its profits to the Treasury, so the Treasury gets $3.1T in exchange for approximately nothing.

Each US citizen then receives a check for $10K. (More realistically, the payments are staggered: perhaps people get ten monthly payments of $1K, or some people get their $10K sooner than others, etc.) The US Mint will need to ramp up production in anticipation of temporary demand to carry more cash around (especially from the unbanked). Also, it'll cost a small amount to print and mail all those checks. However, the primary initial result of sending out all these checks is to flip bits on computers: a person deposits his check at his bank, and some number stored on a computer increments by 10,000.00.

Many people put their $10K into savings and keep it there for years. Most people spend at least some of it soon. Many spend all of it soon. For some things, the price responds to new demand only slowly. For other things, especially globally traded commodities, nominal prices in dollars quickly increase to compensate for the sudden jump in demand by Americans to trade dollars for things. (Increased American demand would also increase the real prices of some goods. This effect is harder to estimate, and could vary a lot by good.)

Nominal wages change very slowly (e.g., I only get a cost-of-living-adjustment once a year). Therefore, in the short term, American real wages decline. Aggregated over all US producers, labor is the biggest cost, so we expect short-term declining real costs for US producers overall. Pair declining domestic producer costs with a sudden increase in domestic demand, and US producers of domestic goods and services will have a big new incentive to expand production.

(There will be international effects too. US exporters have a new cost-side incentive to expand (but no new demand-side incentive). Foreign producers of US imports have a new demand-side incentive to expand (but no new cost-side incentive).)

This expanding production will be self-limiting (in the short term) because it will increase the demand for the inputs to production. This increase will drive up the real prices for many non-labor inputs until it is not profitable to expand any further. [Added: It's not just a supply-side limiting effect; even after a $3.1T shift income, demand still slopes down, so after an initial demand spike increases a price of a good/service, the price would still go down as producers expanded (until further expansion was unprofitable), even if producers did not experience increasing input prices.] However, with 9% unemployment, most US industries will not need to raise real wages to attract new hires. (McDonald's hired 62,000 people last month, out of 1,000,000 applicants!) Also because of high unemployment, for many pre-existing jobs, a real wage increase will not be necessary to retain the employee.

Bottom line: Printing more money right now would be a free lunch for the US economy. The caveat is that this is a statement about an aggregate net gain, not necessarily a net gain for everyone. Fixed-rate mortgage lenders are obvious potential losers, as are bond holders in general. Also, if have you job security, you might be more worried about my talk of declining short-term real wages (long-term real wages are much harder to predict) than about the plight of the unemployed.

It is certainly possible to print too much money. The slack in the labor market is finite; past some point printing more money will just increase all prices, including wages, without increasing production. Moreover, the Fed must credibly commit to keep nominal GDP near its trend level in the long term, not to push it higher. Otherwise, we risk high and unpredictable inflation that will impose real costs by making it hard to plan for the future and diverting business people's time toward more frequent price and wage adjustments. However, we're nowhere close to having too much money. Inflation is below 2%.