That shadowy figure might be me, doing a terrible job.

Photographer: SAUL LOEB/AFP/Getty Images

I Trashed the Economy When I Was Head of the Fed

Mark Gilbert is a Bloomberg View columnist and writes editorials on economics, finance and politics. He was London bureau chief for Bloomberg News and is the author of “Complicit: How Greed and Collusion Made the Credit Crisis Unstoppable.”
Read More.
a | A

The San Francisco branch of the Federal Reserve has a game on its website that lets you play at being Chair of the Federal Reserve. After tinkering with it, I've come to some conclusions: Modeling the economy is a mug's game, short-term interest rates are a poor tool for steering the economy, and I should never be given the job of running a central bank.

The website sets out the objectives:

Your job is to set monetary policy to achieve full employment and low price inflation. Your term will last four years (16 quarters). Keep unemployment close to its natural rate of 5 percent. Keep inflation near the Fed's 2 percent inflation target. Pay attention to the headlines for information about the economy.

The game begins in mid-2020 with the kind of Goldilocks economy that any incoming central bank chief would happily trade their housing allowance for. Inflation is not too hot and not too cold, at 2.1 percent, a level the current Fed hasn't enjoyed for more than two years. Unemployment has been ticking lower for five consecutive quarters and is just 4.7 percent, a smidgen below the 5 percent level that the game says is the "natural" rate and in line with today's 4.9 percent number.

But the benchmark Fed funds rate has remained constant at 4 percent throughout that imagined future period -- a far cry from the current upper bound of 0.5 percent.

So what's an incoming central bank chair to do? Leave well enough alone, of course. Luckily, there are no pesky dissenters to worry about in my monetary dictatorship, no gravitational pull from a zero bound that I'm philosophically anxious to escape, and so my decision to leave rates on hold is unchallenged.

QuickTake Normal Interest Rate

The following quarter sees the jobless rate inch down to 4.6 percent, while inflation speeds up a bit to 2.3 percent. Neither seem particularly worrisome -- because consumer price gains were below the Fed target throughout 2015 and 2016, letting them off the leash for a while seems defensible -- so I'll do what the Fed did last week in the real world and keep rates unchanged again.

"Help Wanted Signs Everywhere," say the resulting news headlines. "Tight job market suggests more inflation ahead." Inflation speeds up to 2.5 percent, while the jobless rate drops to 4.4 percent. Should being so far below the so-called natural jobless rate bother me, the way the labor market seems to currently preoccupy Yellen? Well, in March 2014, the Fed abandoned the idea that a decline below 6.5 percent would be enough to trigger higher borrowing costs. Unemployment has been below 6 percent since October 2014, and yet the Fed has only tweaked its rate higher once, with the current 4.9 percent level not enough to spur action this month.

So, in the game, I stick with that playbook, and keep rates on hold for my third consecutive quarter. In fact, in recognition of the challenge that being "data dependent" is meaningless if you start ignoring the data when it doesn't do what you expect, I'm planning to stand pat on policy all the way until the middle of 2021.

The game tries to interfere with my plans, throwing out what it says is a tax refund for millions of people. (It might instead have had the government taking advantage of stable borrowing costs to boost infrastructure spending, but it seems that much-needed approach is too much to hope for even in fantasy land.) Could this be helicopter money, conjured out of thin air by the central bank rather than coming from existing central government funds? I don't remember authorizing that! The game doesn’t say; but with inflation surging past 3.3 percent and the jobless rate down to close to 3 percent, the economic temperature is definitely rising.

Back in July 2009, when the current Fed chair Janet Yellen was merely head of the San Francisco branch, she said keeping rates near zero for several years was "not outside the realm of possibility." Thus it proved. Yellen took charge of the Fed in Feb. 2014; she didn't oversee her first rate change until December 2015.

So, taking my cue from her, I ignore the game's September 2021 newspaper headlines "Fed Expected to Tighten to Restrain Economy," even as consumer price gains rocket past 4.4 percent.

By the time I'm willing to entertain a rate increase in March 2022 after 20 months at the helm, it's probably too late. Unemployment is 1.5 percent, and has been for months -- the game seems to believe it's impossible for it to drop below that lower bound. Inflation, though, is at a pace not seen since the early 1980s, nudging 7 percent. "Economy Too Hot!" says the newspaper. "Uncontrolled Economic Boom Signals Trouble." The paper is almost definitely preparing my obituary as Fed chief.

Suppose I govern differently from the beginning. It turns out that life's not much better if I start my term by immediately driving the Fed funds rate down to its current 0.5 percent level. Unemployment quickly drops to that 1.5 percent floor, while inflation surges past 4.4 percent. And no matter how many times I click on the "Cut" button, the game won't let me introduce the negative interest rates prevalent in so many of the world's monetary jurisdictions.

In a virtual sense, I'm unique in having served two online terms as Internet Fed chair. I'm no better at it than I was at the start of 2008, when I played the Fed's virtual game using Ben Bernanke's playbook as my guide. Then, a Fed funds level of 1 percent -- back before ZIRP and NIRP were even a gleam in the eyes of unconventional economists -- was enough to drive inflation to double-digit crisis levels. (Soon after the article I wrote then was published, the game disappeared from the Internet. I've only just rediscovered it.) 

The game's changes suggest that the experience of the past decade has changed how economists view the interaction between policy and outcomes. I draw three conclusions from my experience running the Fed in a virtual world.

Firstly, modeling is bunkum, because the real world has so many variables that aren't interacting the way economics textbooks say they should. At no point does the Fed's admittedly simplistic web game tell us what growth is doing, or how wages are performing relative to inflation. Whichever model a central bank employs will always be an imperfect snapshot of the economy. And the growing debate, fueled in part by Fed officials themselves, about whether inflation targets need to be higher, what the true non-accelerating rate of inflation is, and whether targeting gross domestic product makes more sense, illustrates how hard it is to choose what to target.

Secondly, central banks cannot and should not be expected to do all of the work; there's only so much monetary policy can achieve to resuscitate growth without parallel action on the fiscal front from government. The Bank of Japan's decision last week to shift its focus to keeping 10-year yields from dropping below zero percent suggests the typical central bank focus on short-term borrowing costs ignores the effect on the banking industry of longer-term rates.

And thirdly, I'm glad I'm not actually in charge of steering the global economy. I am increasingly sympathetic toward our central bankers; they may have more sophisticated tools than the simulator after embracing unconventional policies, but their jobs are still more difficult than most people imagine.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Mark Gilbert at

To contact the editor responsible for this story:
Therese Raphael at