Job hunters attend a National Career Fair. (Mark Ralston-AFP/Getty Images)

Can Congress or the Federal Reserve fix the economy by buying stuff? Keynesians say yes: Government spending gives people in the private sector more money to spend, which helps businesses and creates jobs. Same deal when the Fed buys up bonds or commits to keep interest rates low. Keynes critics, however, say no: Government spending and Fed action just drive up inflation and make the situation worse than it was before.

A new paper from the Federal Reserve Bank of San Francisco says that, in a way, both sides are correct — but that, in another, more accurate, way, the Keynesians are right.

The key point for understanding this debate is that it all revolves around why the economy is underperforming. If the issue is that people aren't buying things, then fiscal or monetary stimulus would help. If the problem is that businesses just aren't making things, then stimulus could cause inflation.

Enter Sylvain Leduc and Zheng Liu at the San Francisco Fed, who tried to show exactly how the U.S. economy is underperforming. Their method was clever: They looked at economic uncertainty from past shocks to see whether that uncertainty caused both unemployment and prices to rise (which would suggest that businesses aren't making enough stuff) or whether it only triggered only higher unemployment (which would suggest that people are not buying enough stuff).

To separate out the uncertainty caused simply by a weakened economy, Leduc and Liu reviewed consumer and business confidence surveys conducted in a given month before unemployment numbers had been released. The researchers assumed that the responses would be less influenced by official numbers and more likely to reflect the state of the economy.

The authors' model shows that a rise in uncertainty drives up unemployment but lowers inflation, indicating that the uncertainty keeps people from buying things but doesn't necessarily keep businesses from making things:

Source: Sylvain Leduc and Zheng Liu

They also found that uncertainty lowers interest rates, suggesting that fiscal stimulus in the wake of growing uncertainty would not trigger dangerous inflation:

Source: Sylvain Leduc and Zheng Liu

Leduc and Liu's findings suggest that policymakers who want to boost hiring should respond to increases in uncertainty with monetary and fiscal stimulus, just as they would to a fall in aggregate demand.

Their approach, however, has a major flaw. Respondents are still likely to be extremely influenced by the actual state of the economy, because of their personal experience and by having seen the previous month's official numbers. Given that monthly employment data doesn't typically vary much, it seems quite unlikely that a person who is even four to five weeks removed from the most recent data would not be influenced by the actual state of the economy.

But even if Leduc and Liu's method failed to differentiate between uncertainty on its own and uncertainty caused by underlying economic conditions, the takeaway is the same. If a rise in uncertainty is caused, ultimately, by weak demand, then uncertainty surveys are a measure of the state of demand, and so increases in uncertainty suggest that demand, not supply, is falling. So easing measures like QE3 will raise both aggregate demand and reduce uncertainty, and aid the recovery in the process.

So, the next time you hear a politician or economist say that uncertainty, not demand, is the problem, know that it can be both!