Tuesday, July 8, 2014

A Legislated Taylor Rule?

John Taylor announces in his blog post, "New Legislation Requires Fed to Adopt Policy Rule'' that today (June 8)

.. the ‘‘Federal Reserve Accountability and Transparency Act of 2014” was introduced into Congress. It requires that the Fed adopt a rules-based policy. Basically, the Fed would have to report to Congress and explain any deviation from a "Reference policy rule,"
The term ‘Reference Policy Rule’ means a calculation of the nominal Federal funds rate as equal to the sum of the following: (A) The rate of inflation over the previous four quarters. (B) One-half of the percentage deviation of the real GDP from an estimate of potential GDP. (C) One-half of the difference between the rate of inflation over the previous four quarters and two. (D) Two.
Wow. John will testify at a hearing at the House Financial Service Committee on Thursday, along with Mark Calabria, Hester Peirce and Simon Johnson. This should be very interesting.

Questions for discussion:

1. What is most interesting about a rule is what it leaves out. Notably absent here is "macroprudential" policy, "financial stability" goals, i.e. raising rates to prick perceived asset price "bubbles" and so forth. Janet Yellen's remarkable recent speech foreswore a lot of that.

Of course, the Fed could always add it as a "temporary" need to deviate from the rule. Still, many people might think that should be part of the rule not part of the exception. It also leaves out housing, exchange rates, and all the other things that central banks like to pay attention to.

A rule really is a list of things that the Fed shall not react to without explanation.

2. I will be interested to hear the debate between inflation targeters and rules advocates. Inflation targeting is a similar legislative approach, but it basically says "here is the goal. It is a very limited goal. Don't pay attention to anything else, and we won't blame you for anything else. Do whatever you want to get there." A rule prescribes the actions the Fed should take, with only limited statement about what the goal should be.

Inflation targeting is like "go to Minneapolis, not St. Louis, and don't get distracted by shopping along the way. The rest is up to you, wake me up when we're there." A rule is like "Stay on I-94. When the white line gets too close to the right wheels, turn a bit to the left; when the dashed line gets too close to the left wheels, turn a bit to the right. If you need to go to the bathroom, wake me up and tell me why we're getting off the freeway."

I am making too light of it, as these are serious issues. The point is to enhance the stability and predictability of monetary policy, to "anchor expectations," to help the Fed to precommit ex ante to actions it will be tempted to take ex-post, and to help the Treasury to precommit ex ante to provide the fiscal and legislattive support necessary to fight inflation or deflation -- an often overlooked issue -- and to precommit that Congress will not complain about the Fed if it follows the rule.

Rules vs. targets is a deep question that needs to consider political economy, expectations, game theory, and so forth.

3. Of course, if this goes anywhere we will have a big debate over what rule to enshrine in Federal legislation for a generation. Price level target or inflation rate target? Two? What about secular stagnation? Expected inflation or past inflation? Oh-Oh, here come the nominal GDP targeters. And the monetarists...

4. This only covers the short term Federal Funds rate, which may remain at zero for decades the way things are going. What about rules for asset purchases, macroprudential regulation, etc. etc.?


  1. John,

    I am all for rules base monetary policy. However, rules based monetary policy will fail without rules based fiscal policy - see Euro.

    Fed wants to set short term interest rate at rule based level, Congress wants to spend without raising taxes to pay for it. Congress already loaded down with debt faces cash flow risk when rates rise - coupon payments on debt can exceed available tax revenue.

    Until Mr. Taylor is willing to include rules based fiscal policy (balanced budget amendment, no bailouts, government equity, zero coupon bonds, etc.) in his proposed legislation, it is dead in the water.

    1. Good point Frank. I’d go further. First, having two entirely separate bodies influencing aggregate demand (the Fed and Congress) is daft. It’s like having a car with two steering wheels each controlled by a husband and wife in the middle of a matrimonial breakdown.

      Second, why adjust AD by trying to adjust just ONE type of economic activity: that’s lending based types of activity (mainly investment), which is what interest rate adjustments attempt to do. One might as well adjust AD by trying to increase just car, clothes and ice-cream sales.

    2. Ralph,

      Not all problems are aggregate demand problems. There can be problems with the supply side, there can be problems with cronyism and/or monopolistic practices, and there can be problems with fraud as well.

      How about this instead - central bank worries about aggregate demand, government worries about aggregate supply - aka separation of powers?

      My point was that to separate those powers (aggregate demand management / aggregate supply management), the federal government needs to get out of the borrowing business. Meaning changes in interest rates set by the central bank do not affect the federal government's cash flow position at all.

      When you are trying to manage both aggregate supply and demand, you need two cars, two steering wheels, and two drivers.

  2. Rules are good for determining any policy. If one makes a rule or rules for monetary policy, it should include a real component, such as real GDP growth or the unemployment rate, for obvious, old fashioned, reasons, such as the low nominal interest rates prevailing during the Great Depression. In case of financial crisis, "discount freely"?

  3. A problem with this particular proposal is that the "rule" depends on "an estimate of potential GDP". Imagine two regimes, one which does whatever it wants and adjusts its estimate of potential GDP to make its decisions conform to the rule, and another which makes an "honest effort" to estimate potential GDP and then follows the rule accordingly. How could we possibly tell the difference between the two regimes (as long as what the policymakers want is not too unusual)?

    The advantage of a target is that we can actually assess performance objectively. E.g., the inflation target is 2 percent, but inflation has been running at 1% for a year--bad. Unless the rule entirely consists of observable inputs, it seems the target wins this contest handily.

    1. I most agree, but remember, even estimates of inflation are subjective, and sometimes misleading. There are some who think even the PCE deflation overestimates inflation, due to rapid evolutions in products, services and retailing.

      I think the right monetary policy keeps aggregate real demand growing, somewhere north of 3 percent a year, or nominal demand around 6 percent a year.

    2. I'm putting the question of which variable to target to one side for now (though targeting real variables is a recipe for price level indeterminacy). What I wonder is why anyone should care about the means as opposed to the end. Unless you think the certain routes to hitting the target have negative side effects relative to other routes, it would seem that hitting the target, whatever the target is, is the goal, not following a certain route wherever it happens to lead.

  4. I happen to favor a rules-based policy, although in my case a Market Monetarist approach, and run the engines a little hot. I like prosperity, a dirty word in central bankerly circles. If we have mild inflation, so be it.

    That said, I like some sort of rules and Taylor is a smart guy.

    But this caveat sort of pulls the rug out: The Fed must "explain any deviation" from the rules under ‘Federal Reserve Accountability and Transparency Act of 2014.'

    You mean, the Fed muffs it or goes off policy, so they just go up on the Hill and they "explain any deviation." The Fed has teams of writers and economists to make any explanation sound shrewd, prudent and intelligent. Sad to say, most Congressmen are either uninterested, or they are gold-nut, inflation-hysteric table-thumpers (whatever happened to the silver standard, anyway?).

    It has been noted that central banks will never stop asphyxiating an economy until there is political arm-twisting or takeover from above. See Japan and Abenomics (which seems to be working, btw).

    Another strange fact: For 30 years in the Western developed economies., inflation and interest rates have been in a secular swoon. We are now hitting ZLB in Japan and Europe, and USA maybe next.

    But what has the table-thumpers in hysterics?

    Inflation! Hyperinflation! Galloping inflation!

    Don't you think the shoe is on the other foot?

  5. It seems to me that the rule, if followed slavishly, would produce a very delayed policy response in e.g. a crisis situation such as that in September 2008. I haven't looked up the exact figures, but - if memory serves correctly - inflation was fairly high from September 2007 to September 2008, and the big drop in GDP only came after September 2008. The implication would be that the Fed, rather than lowering rates as much as possible, should keep them fairly high.

  6. Soren:

    Excellent point!

    The Fed in 2008 responded to commodities inflation. A very bad idea. The price of oil set in global markets and the price of gold set in India and China. Corn had an ethanol shot. The Fed was fighting global commodities inflation with US monetary policy, like it was 1964 or something.

    Also, as I read the above formula, if we had no inflation, and no economic growth, the federal funds rate would be 4.5 percent. Seems kinda high.

    Now it might be (boy this is terrible wording) that (B) means negative 1.5 percent in the case of a zero growth economy. In that case, the fed funds rate is 0.5 percent..

    B) One-half of the percentage deviation of the real GDP from an estimate of potential GDP [is this positive or negative in a zero growth economy?]

    In a deflationary recession, I guess the Fed Funds rate would have to fall into negative territory according to the above rules.

  7. Is there a rule that would have prevented the 2008 crash?

    1. Yes. Thou shalt not fund illiquid risky assets by overnight borrowing at 30:1 leverage. See "toward a run free financial system"

    2. Milton Friedman also advocated Prof.Cochrane’s “thou shalt not…” idea. See Ch3 of Friedman's book “A Program for Monetary Stability”, starting at section entitled “Banking Reform”.

  8. "The term ‘Reference Policy Rule’ means a calculation of the nominal Federal funds rate as equal to the sum of the following: (A) The rate of inflation over the previous four quarters. (B) One-half of the percentage deviation of the real GDP from an estimate of potential GDP. (C) One-half of the difference between the rate of inflation over the previous four quarters and two. (D) Two."

    Okay, imagine we are in a deflationary recession, both economic growth and inflation at minus 2 percent. Maybe like 2008.

    a= -2
    b=2.5 (but is this a negative 2.5 or positive? I think negative, but they need to make this clear. I am assuming real potential growth of 3 percent).
    c=2 but again it probably is minus or negative 2.

    In a deflationary recession, I think this rule calls for a fed funds rate of negative 4.5 percent.

    My guess is that the Fed would have little problem explaining why the fed funds rate was not -4.5 percent.

    1. Ben,

      This is why rule based monetary policy will not work without rule based fiscal policy.

      A = Rate of inflation over the past four quarters
      B1 = Present Real GDP growth rate
      B2 = Potential Real GDP growth rate

      Set monetary policy tool = B2
      Set fiscal policy tool = B2 - B1

      Here, monetary policy does not act unless there is some force that destroys real potential growth.

      In a credit based monetary economy, fiscal policy is used to offset monetary policy. Simple examples are the variety of interest deductions (mortgage, student loan, business) embedded into the U. S. tax code.

      And so in your example, even in a deflationary recession, fiscal policy overcomes the gap between real and potential real GDP growth.

      The problem with the various interest deductions in the U. S. tax code are that they only account for nominal interest cost instead of real interest cost. If nominal rates are 3% but real rates are 7%, you can only deduct the nominal 3% interest on a loan.

    2. Frank--thanks for your comment. But for now, as is, advocates for monetary rules need to explain what happens when their rules call for negative interest rates. I just showed that in a deflationary recession, the rules discussed call for a negative 4.5 percent federal funds rate.
      Better are rules that focus on targets, not instruments. I hope Cochrane addresses this.

    3. Ben,

      Economic targets were already set by Congress here:


      1. Explicitly states that the federal government will rely primarily on private enterprise to achieve the four goals. (Which stimulus fans want to override).

      2. Instructs the government to take reasonable means to balance the budget. (Which they continually fail to do).

      3. Instructs the government to establish a balance of trade, i.e., to avoid trade surpluses or deficits. (Which they continually fail to do).

      3. Mandates the Board of Governors of the Federal Reserve to establish a monetary policy that maintains long-run growth, minimizes inflation, and promotes price stability. (They had a pretty good record until recently)

      "Better are rules that focus on targets, not instruments."

      The central bank needs an instrument to reach a target. John did address this. His argument was that a rule focused solely on a target with no mention of the instrument used isn't a "rule" at all.

      It would be like the Chicago Cubs management instituting a "rule" that the Cubs will win the World Series with no mention of how that objective is reached. Humphrey Hawkins suffers some of the same criticism (though balance of budget should be self-evident).

  9. Well one thing is for certain: the Fed has never followed anything resembling a Taylor Rule in the past: http://www.separatinghyperplanes.com/2014/06/how-has-taylor-rule-performed.html

    1. The Taylor rule is claimed by some to be an ad-hoc representation of the Greenspan led Fed. Note that the "rule based" changes in interest rates by the Greenspan Fed in 1994-1995 were proceeded by the federal government making an earnest attempt to balance its books in 1993 -


      Had the 1993 budget not passed, it is likely that short term interest rates set by the Greenspan Fed would have been placed on a different trajectory.

      Greenspan needed Congress to get its act together before he could justify higher interest rates. Sound familiar?

  10. All considered, I mildly support the rule only because it will bring at least an illusion of stability and predictability. And that will be good for markets.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.