I use the term "textbook macro" a lot. I want to stop and define it.
In textbook macro, sometimes there is excess supply of labor and sometimes there is not. That is the supply side of the economy. The demand side of the economy has two policy tools: monetary policy, where an expansionary move is when the Fed buys bonds with money that it creates; fiscal policy, where an expansionary move is a spending increase or tax cut that raises the deficit.
(a) When there is little or no excess labor, an expansionary policy results in little or no change to real wages or employment.
(b) When there is a lot of excess labor, an expansionary policy results in a small but significant increase in prices, which results in a significant drop in real wages, which in turn causes a significant increase in employment.
The overall inflation effect is smaller in (b) than in (a). Instead, in (b) you get a lot more output with only a little more inflation.
Textbook monetarists believe that monetary expansions are necessary and sufficient to raise prices. Of course, when there is excess labor, once prices go up and wages are sticky, we get lower real wages and more employment.
Textbook Keynesians believe that monetary expansions can be ineffective. One story is that a monetary expansion will not do much to interest rates. Another story is that reducing interest rates will not do much to increase investment. Either way, overall spending will not change much, and it is as if the money gets put under a mattress, so prices do not go up. For that, you need fiscal expansion.
Next, we get to Scott Sumner's macro, which is textbook macro with a slight twist. He writes,
Suppose we target NGDP growth over the next 5 years, surely all wages and prices would have adjusted by then? Yes, but the point of changing future expected NGDP is to change current AD, and hence current NGDP. And if we do that, and if wages and prices are sticky, then monetary policy can have real effects in the short run.
Here is an analogy. Suppose than the US government announced that starting three years from today it would begin buying unlimited quantities of gold at $2000/ounce. But for now it would not intervene in the gold market, leaving prices at their current $1250. Obviously the policy, if credible, would cause gold prices to immediately rise to about $1900, and employment in the gold mining industry would rise.
The gold story is easy to follow. If we all work in the GDP factory (or gold mine), then if the Fed sets a credible target for the price of GDP (or gold) and the workers are stupid enough to leave wages set without regard to what that target does to the price of GDP (or gold) today, then by golly real wages will fall and we will get more employment.
I have two problems with Sumner's story.
1. Money and Nominal Targets
My first problem is that I do not think that the Fed can hit precise nominal targets. Instead, I think that the Fed can aim for one of two regimes. In the first regime, inflation is low and stable. In that regime, people are relatively indifferent about what the Fed is up to. Long-term interest rates are set in the bond market. Unless the Fed wants to go beserk, it has to set short-term rates within a range consistent with those expectations. But only a few Wall Street securities traders who deal in short-term instruments have a big stake in what the Fed does day to day or month to month.
In the second regime, inflation is high and variable. In that regime, people have lost their faith in the predictability of the value of money. They do everything they can to economize on cash holdings, and those efforts in turn raise the velocity of money and make velocity unstable. A lot of real resources are being used up by people trying to evade the inflation tax.
In the first regime, an announcement of a target for nominal GDP would be met with indifference. The Fed does not have credibility in achieving a nominal GDP target. As long as the Fed does not go beserk, the changes that it makes in its balance sheet will hardly alter anyone's behavior. The first regime looks like the textbook Keynesian case (although if you think I am a textbook Keynesian, keep reading).
In the second regime, the Fed cannot hit a nominal GDP target because inflation and velocity are moving around too quickly. At best, the Fed can commit to aiming to get back to the first regime, in which case it probably will take time to establish credibility.
I have in mind that we were in the first regime in the 1950's and 1960's, got into the second regime in the 1970's (distorted somewhat by wage and price controls from 1971-1973), and got back to the first regime starting with a commitment by Paul Volcker in 1980 that the markets did not really fully believe until the mid-1990's.
2. The Nature of Unemployment
Here is where Sumner relies on textbook macro, and I reject textbook macro, which says that all you have to do to get rid of unemployment is sprinkle lower real wages on workers, and poof!
I think I expressed myself well in the post What is Job Creation? Nowadays, jobs are like investments. I would say that late in 2008 and early in 2009 firms decided on net to make unusually steep disinvestments. Why did they do this? Sumner says that it is because business leaders suddenly decided that the Fed was aiming for lower nominal GDP. I do not find that story persuasive. I think that business leaders became very focused on short-term survival, in many cases with justification.
Nowadays, as a business leader you look at your balance sheet, assess the outlook, and make your investment decisions. You hire workers when you see opportunities, and you fire workers when you face threats. The ratio of threats to opportunities went way up in the latter half of 2008. I am skeptical that monetary or fiscal policy can be used to improve that ratio.