(Reuters) - Janet Yellen owes part of her dovish reputation to her embrace of an "optimal control" policy approach that tolerates inflation temporarily overshooting the Federal Reserve's 2 percent goal in return for a quicker decline in U.S. unemployment.
Investors are likely to weigh more carefully what such an approach might mean for U.S. monetary policy after she was confirmed by the Senate on Monday to succeed Ben Bernanke at the helm of the central bank.
Economists at Goldman Sachs think optimal control considerations mean the central bank is likely to hold off raising interest rates for longer than many financial market participants expect.
Might a Yellen Fed raise interest rates too slowly and allow inflation to get out of control? Or is the bigger risk that it tightens policy prematurely and inflicts another recession?
Optimal policy -- a technical term economists use to describe maximizing good outcomes -- will help frame that debate.
Yellen first talked publicly about an optimal control path for rates in a June 2012 speech. But the idea of a mathematical model to summarize policy trade-offs -- in this case between unemployment and inflation -- is common to economics.
It describes a way to set policy that optimizes economic outcomes according to the Fed's dual mandate for price stability and maximum sustainable employment.
As described by Yellen, the Fed could achieve a faster reduction in U.S. unemployment at the modest price of a bit more temporary inflation than it would get if it set policy with rigid respect for its 2 percent inflation goal.
"I assume that monetary policy aims to minimize the deviations of inflation from 2 percent and the deviations of the unemployment rate from 6 percent, with equal weight on both objectives," she said in November 2012. At the time, 6 percent was the upper edge of the range of jobless rates that the Fed felt was in keeping with full employment.
When she ran the numbers for that speech, the model suggested the Fed should hold rates near zero until early 2016, and keep them lower than recommended by a more traditional policy rule until 2018.
The payoff would be unemployment falling to 6.5 percent by mid-2015, versus 7 percent under what was then the Fed's baseline forecast, and 7.5 percent if the Fed followed the more traditional rule that recommended earlier rate hikes.
Skeptics say the optimal control approach may optimize economic outcomes, but it lacks control and increases the risk of a policy error.
"The optimal control framework actually provides the illusion of control without necessarily control itself," said Carl Tannenbaum, chief economist for Northern Trust in Chicago.
By tolerating higher inflation to achieve its employment goal, critics worry the Fed might not realize it has a problem until it is too late.
Further, they worry that the inflation and unemployment goals are too loosely defined to keep the central bank on the straight and narrow. The Fed targets 2 percent inflation, but shoots for that goal over the "medium term."
As for the jobless rate it tries to achieve, officials now say it is likely between 5.2 percent and 5.8 percent, but they don't know for sure and admit it could change over time.
"You might end up with the Fed probably being able to justify a broad range of actions on the basis of the same set of data," said Tannenbaum. "It may seem more scientific than it actually is."
Economists who think Yellen is on the right track prefer to think about the optimal control approach as a "common sense" way to describe a problem faced by the Fed when inflation is lower than desired, as is currently the case.
"Optimal control is just language for pointing out that in that case, it is a much clearer policy choice, if you think inflation at the moment is too low and ... unemployment is too high," said James Hamilton, an economics professor at the University of California, San Diego.
Also, the risk of a policy mistake cuts both ways, and a decision to tighten policy too soon might mean hundreds of thousands, if not millions, of Americans being without work for longer than would have otherwise been the case.
"You have to acknowledge that may be having a permanent effect on people's lives, and for that matter a permanent effect on what we can accomplish as a nation," Hamilton said.
Yellen has said the model is instructive and a useful guidepost, but that it's no replacement for judgment. She has said she would "never advocate turning over monetary policy to a computer."
The Fed, however, has taken steps that are very much in keeping with the spirit behind an optimal control approach.
To convince financial markets they will keep rates low for a long time, officials had said they would not raise rates at least until unemployment hit 6.5 percent, provided the outlook for inflation stayed under 2.5 percent.
On December 18 the Fed tweaked that pledge, saying it likely would be appropriate to keep rates near zero "well past the time" that the jobless rate fell below 6.5 percent, especially if inflation expectations remained below target.
The U.S. jobless rate in November was 7.0 percent.
Reporting by Washington economics team; Editing by Paul Simao and Chizu Nomiyama