Cluelessness about Strategy, Tactics and Discretion

In his op-ed in the weekend Wall Street Journal, John Taylor restates his confused opposition to what Ben Bernanke calls the policy of constrained discretion followed by the Federal Reserve during his tenure at the Fed, as vice-chairman under Alan Greenspan from 2003 to 2005 and as Chairman from 2005 to 2013. Taylor has been arguing for the Fed to adopt what he calls the “rules-based monetary policy” supposedly practiced by the Fed while Paul Volcker was chairman (at least from 1981 onwards) and for most of Alan Greenspan’s tenure until 2003 when, according to Taylor, the Fed abandoned the “rules-based monetary rule” that it had followed since 1981. In a recent post, I explained why Taylor’s description of Fed policy under Volcker was historically inaccurate and why his critique of recent Fed policy is both historically inaccurate and conceptually incoherent.

Taylor denies that his steady refrain calling for a “rules-based policy” (i.e., the implementation of some version of his beloved Taylor Rule) is intended “to chain the Fed to an algebraic formula;” he just thinks that the Fed needs “an explicit strategy for setting the instruments” of monetary policy. Now I agree that one ought not to set a policy goal without a strategy for achieving the goal, but Taylor is saying that he wants to go far beyond a strategy for achieving a policy goal; he wants a strategy for setting instruments of monetary policy, which seems like an obvious confusion between strategy and tactics, ends and means.

Instruments are the means by which a policy is implemented. Setting a policy goal can be considered a strategic decision; setting a policy instrument a tactical decision. But Taylor is saying that the Fed should have a strategy for setting the instruments with which it implements its strategic policy.  (OED, “instrument – 1. A thing used in or for performing an action: a means. . . . 5. A tool, an implement, esp. one used for delicate or scientific work.”) This is very confused.

Let’s be very specific. The Fed, for better or for worse – I think for worse — has made a strategic decision to set a 2% inflation target. Taylor does not say whether he supports the 2% target; his criticism is that the Fed is not setting the instrument – the Fed Funds rate – that it uses to hit the 2% target in accordance with the Taylor rule. He regards the failure to set the Fed Funds rate in accordance with the Taylor rule as a departure from a rules-based policy. But the Fed has continually undershot its 2% inflation target for the past three years. So the question naturally arises: if the Fed had raised the Fed Funds rate to the level prescribed by the Taylor rule, would the Fed have succeeded in hitting its inflation target? If Taylor thinks that a higher Fed Funds rate than has prevailed since 2012 would have led to higher inflation than we experienced, then there is something very wrong with the Taylor rule, because, under the Taylor rule, the Fed Funds rate is positively related to the difference between the actual inflation rate and the target rate. If a Fed Funds rate higher than the rate set for the past three years would have led, as the Taylor rule implies, to lower inflation than we experienced, following the Taylor rule would have meant disregarding the Fed’s own inflation target. How is that consistent with a rules-based policy?

It is worth noting that the practice of defining a rule in terms of a policy instrument rather than in terms of a policy goal did not originate with John Taylor; it goes back to Milton Friedman who somehow convinced a generation of monetary economists that the optimal policy for the Fed would be to target the rate of growth of the money supply at a k-percent annual rate. I have devoted other posts to explaining the absurdity of Friedman’s rule, but the point that I want to emphasize now is that Friedman, for complicated reasons which I think (but am not sure) that I understand, convinced himself that (classical) liberal principles require that governments and government agencies exercise their powers only in accordance with explicit and general rules that preclude or minimize the exercise of discretion by the relevant authorities.

Friedman’s confusions about his k-percent rule were deep and comprehensive, as a quick perusal of Friedman’s chapter 3 in Capitalism and Freedom, “The Control of Money,” amply demonstrates. In practice, the historical gold standard was a mixture of gold coins and privately issued banknotes and deposits as well as government banknotes that did not function particularly well, requiring frequent and significant government intervention. Unlike, a pure gold currency in which, given the high cost of extracting gold from the ground, the quantity of gold money would change only gradually, a mixed system of gold coin and banknotes and deposits was subject to large and destabilizing fluctuations in quantity. So, in Friedman’s estimation, the liberal solution was to design a monetary system such that the quantity of money would expand at a slow and steady rate, providing the best of all possible worlds: the stability of a pure gold standard and the minimal resource cost of a paper currency. In making this argument, as I have shown in an earlier post, Friedman displayed a basic misunderstanding of what constituted the gold standard as it was historically practiced, especially during its heyday from about 1880 to the outbreak of World War I, believing that the crucial characteristic of the gold standard was the limitation that it imposed on the quantity of money, when in fact the key characteristic of the gold standard is that it forces the value of money – regardless of its material content — to be equal to the value of a specified quantity of gold. (This misunderstanding – the focus on control of the quantity of money as the key task of monetary policy — led to Friedman’s policy instrumentalism – i.e., setting a policy rule in terms of the quantity of money.)

Because Friedman wanted to convince his friends in the Mont Pelerin Society (his egregious paper “Real and Pseudo Gold Standards” was originally presented at a meeting of the Mont Pelerin Society), who largely favored the gold standard, that (classical) liberal principles did not necessarily entail restoration of the gold standard, he emphasized a distinction between what he called the objectives of monetary policy and the instruments of monetary policy. In fact, in the classical discussion of the issue by Friedman’s teacher at Chicago, Henry Simons, in an essay called “Rules versus Authorities in Monetary Policy,” Simons also tried to formulate a rule that would be entirely automatic, operating insofar as possible in a mechanical fashion, even considering the option of stabilizing the quantity of money. But Simons correctly understood that any operational definition of money is necessarily arbitrary, meaning that there will always be a bright line between what is money under the definition and what is not money, even though the practical difference between what is on one side of the line and what is on the other will be slight. Thus, the existence of near-moneys would make control of any monetary aggregate a futile exercise. Simons therefore defined a monetary rule in terms of an objective of monetary policy: stabilizing the price level. Friedman did not want to settle for such a rule, because he understood that stabilizing the price level has its own ambiguities, there being many ways to measure the price level as well as theoretical problems in constructing index numbers (the composition and weights assigned to components of the index being subject to constant change) that make any price index inexact. Given Friedman’s objective — demonstrating that there is a preferable alternative to the gold standard evaluated in terms of (classical) liberal principles – a price-level rule lacked the automatism that Friedman felt was necessary to trump the gold standard as a monetary rule.

Friedman therefore made his case for a monetary rule in terms of the quantity of money, ignoring Simons powerful arguments against trying to control the quantity of money, stating the rule in general terms and treating the selection of an operational definition of money as a mere detail. Here is how Friedman put it:

If a rule is to be legislated, what rule should it be? The rule that has most frequently been suggested by people of a generally liberal persuasion is a price level rule; namely, a legislative directive to the monetary authorities that they maintain a stable price level. I think this is the wrong kind of a rule [my emphasis]. It is the wrong kind of a rule because it is in terms of objectives that the monetary authorities do not have the clear and direct power to achieve by their own actions. It consequently raises the problem of dispersing responsibilities and leaving the authorities too much leeway.

As an aside, I note that Friedman provided no explanation of why such a rule would disperse responsibilities. Who besides the monetary authority did Friedman think would have responsibility for controlling the price level under such a rule? Whether such a rule would give the monetary authorities “too much leeway” is of course an entirely different question.

There is unquestionably a close connection between monetary actions and the price level. But the connection is not so close, so invariable, or so direct that the objective of achieving a stable price level is an appropriate guide to the day-to-day activities of the authorities. (p. 53)

Friedman continues:

In the present state of our knowledge, it seems to me desirable to state the rule in terms of the behavior of the stock of money. My choice at the moment would be a legislated rule instructing the monetary authority to achieve a specified rate of growth in the stock of money. For this purpose, I would define the stock of money as including currency outside commercial banks plus all deposits of commercial banks. I would specify that the Reserve System shall see to it [Friedman’s being really specific there, isn’t he?] that the total stock of money so defined rises month by month, and indeed, so far as possible day by day, at an annual rate of X per cent, where X is some number between 3 and 5. (p. 54)

Friedman, of course, deliberately ignored, or, more likely, simply did not understand, that the quantity of deposits created by the banking system, under whatever definition, is no more under the control of the Fed than the price level. So the whole premise of Friedman’s money supply rule – that it was formulated in terms of an instrument under the immediate control of the monetary authority — was based on the fallacy that quantity of money is an instrument that the monetary authority is able to control at will.

I therefore note, as a further aside, that in his latest Wall Street Journal op-ed, Taylor responded to Bernanke’s observation that the Taylor rule becomes inoperative when the rule implies an interest-rate target below zero. Taylor disagrees:

The zero bound is not a new problem. Policy rule design research took that into account decades ago. The default was to move to a stable money growth regime not to massive asset purchases.

Taylor may regard the stable money growth regime as an acceptable default rule when the Taylor rule is sidelined at the zero lower bound. But if so, he is caught in a trap of his own making, because, whether he admits it or not, the quantity of money, unlike the Fed Funds rate, is not an instrument under the direct control of the Fed. If Taylor rejects an inflation target as a monetary rule, because it grants too much discretion to the monetary authority, then he must also reject a stable money growth rule, because it allows at least as much discretion as does an inflation target. Indeed, if the past 35 years have shown us anything it is that the Fed has much more control over the price level and the rate of inflation than it has over the quantity of money, however defined.

This post is already too long, but I think that it’s important to say something about discretion, which was such a bugaboo for Friedman, and remains one for Taylor. But the concept of discretion is not as simple as it is often made out to be, especially by Friedman and Taylor, and if you are careful to pay attention to what the word means in ordinary usage, you will see that discretion does not necessarily, or usually, refer to an unchecked authority to act as one pleases. Rather it suggests that a certain authority to make a decision is being granted to a person or an official, but the decision is to be made in light of certain criteria or principles that, while not fully explicit, still inform and constrain the decision.

The best analysis of what is meant by discretion that I know of is by Ronald Dworkin in his classic essay “Is Law a System of Rules?” Dworkin discusses the meaning of discretion in the context of a judge deciding a “hard case,” a case in which conflicting rules of law seem to be applicable, or a case in which none of the relevant rules seems to fit the facts of the case. Such a judge is said to exercise discretion, because his decision is not straightforwardly determined by the existing set of legal rules. Legal positivists, against whom Dworkin was arguing, would say that the judge is able, and called upon, to exercise his discretion in deciding the case, meaning, that by deciding the case, the judge is simply imposing his will. It is something like the positivist view that underlies Friedman’s intolerance for discretion.

Countering the positivist view, Dworkin considers the example of a sergeant ordered by his lieutenant to take his five most experienced soldiers on patrol, and reflects on how to interpret an observer’s statement about the orders: “the orders left the sergeant a great deal of discretion.” It is clear that, in carrying out his orders, the sergeant is called upon to exercise his judgment, because he is not given a metric for measuring the experience of his soldiers. But that does not mean that when he chooses five soldiers to go on patrol, he is engaging in an exercise of will. The decision can be carried out with good judgment or with bad judgment, but it is an exercise of judgment, not will, just as a judge, in deciding a hard case, is exercising his judgment, on a more sophisticated level to be sure than the sergeant choosing soldiers, not just indulging his preferences.

If the Fed is committed to an inflation target, then, by choosing a setting for its instrumental target, the Fed Funds rate, the Fed is exercising judgment in light of its policy goals. That exercise of judgment in pursuit of a policy goal is very different from the arbitrary behavior of the Fed in the 1970s when its decisions were taken with no clear price-level or inflation target and with no clear responsibility for hitting the target.

Ben Bernanke has described the monetary regime in which the Fed’s decisions are governed by an explicit inflation target and a subordinate commitment to full employment as one of “constrained discretion.” When using this term, Taylor always encloses it in quotations markets, apparently to suggest that the term is an oxymoron. But that is yet another mistake; “constrained discretion” is no oxymoron. Indeed, it is a pleonasm, the exercise of discretion usually being understood to mean not an unconstrained exercise of will, but an exercise of judgment in the light of relevant goals, policies, and principles.

PS I apologize for not having responded to comments recently. I will try to catch up later this week.

15 Responses to “Cluelessness about Strategy, Tactics and Discretion”

  1. 1 Frank Restly May 6, 2015 at 9:03 am


    One of your best articles, thanks.

    A crucial element Taylor misses is causality between instrument (Fed funds rate) and objective (inflation rate). He relies on past history to establish that causality, but cherry picks his period of history (1981-2003). Going back a little further in history (say 1970-1980), it should be apparent that increases in the inflation rate were often proceded in time by increases in the Fed funds rate.

    I would not argue direct causality between the two (inflation rate and fed funds rate) because there are too many other variables in play (fiscal stance being a big one), but you have to wonder if Taylor has his signs wrong when doing policy prescriptions.

  2. 2 Philo May 7, 2015 at 9:27 am

    This is well said: “the exercise of discretion [is] usually understood to mean not an unconstrained exercise of will, but an exercise of judgment in the light of relevant goals, policies, and principles.” And it would be unrealistic to try *utterly to eliminate* a public official’s exercise of discretion in the discharge of his duties. Still, it is reasonable to want to *reduce* the monetary authority’s exercise of discretion *to a minimum*, with the idea of making the course of monetary policy more reliable and predictable. With discretion, *poor judgment* is an ever-present possibility, and a general *fear* that the monetary authority’s judgment will be poor (or even simple uncertainty about what, specifically, it will be) may detract from economic stability. Other things equal, a rule that leaves less room for discretion is better.

    Dworkin’s lieutenant might have ordered the sergeant to take his five *tallest* men; that would have left the sergeant with less discretion than the actual order to take the five *most experienced*. At an extreme, he might have specified which men were to be taken; that would not *necessarily* have eliminated the sergeant’s discretion about whom to take, since it is impossible to include explicit provisos for every possible eventuality–suppose, for example, one of the specified men fell ill just before the operation, etc.–but it *might* have done so. On the other hand, the lieutenant might have ordered the sergeant to take the five *best* men; that would have left him more discretion. At an extreme, he might have ordered him to take *whichever group of men, of whatever size and composition, he expected would produce the best overall results on the whole, in the long run*; that would still impose a constraint on the sergeant’s choice, but one that left him enormous scope for the exercise of discretion.

    In short, discretion is a matter of degree. A lieutenant may do well to trust his sergeant’s judgment, but I have minimal trust in the judgment of the present, and any likely future, FOMC. I wish their discretion were reduced to the practicable minimum.

  3. 3 David Glasner May 8, 2015 at 11:07 am

    Frank, In the 1970s there were supply shocks that increased inflation, that doesn’t prove that inflation would not have been even higher if the Fed Funds rate had not been raised, which is not to say that raising the interest rate after a supply shock was the correct policy.

    Philo, I certainly agree that discretion is a matter of degree, but it’s not as if there is any reliable way to establish what is the minimum amount of discretion to hand over to the authorities. That is an underlying fallacy in the rules versus discretion debate — there are many others, Judges have discretion to decide hard cases. Does that mean the rule of law is an illusion? Not necessarily. Unfortunately, to listen Friedman and Taylor, you would think that any exercise of judgment by the monetary authority is an outrage against the rule of law. That’s just stupid. And their proposed rules-based policy of setting an instrument of monetary policy (or, in Friedman’s case a pseudo-instrument) in accordance with some “rule” that Taylor wrote down on the back of an envelope actually subverts the policy rule that most reasonable people would think ought to take precedence over a rule for setting an instrument.

  4. 4 Philo May 11, 2015 at 9:06 am

    I did not say there was a “reliable way to establish what is the minimum amount of discretion to hand over to the authorities”; I simply said that *ceteris paribus* the less discretion the better. Friedman and Taylor may present the alternatives as white and black, when really they are various shades of gray, but the kernel of validity in their view is that we should tend to view lighter shades of gray more favorably than darker.

  5. 5 David Glasner May 11, 2015 at 9:37 am

    Philo, Not sure about that at all. There is a difference in kind between a rule formulated in terms of an objective for monetary policy, e.g., price level, price of a commodity (e.g., gold), a rate of inflation, and a rule formulated in terms of an instrument. I think instrument rules are dangerous, because there is no reason to assume that the relationship between the instrument and the target can be usefully described in terms of a rule that could be implemented. And of course there is also the problem that what someone thinks is an instrument turns out to be a target, and it may be a bad target. And does the maxim of reducing discretion wherever possible really give us any useful guidance in choosing between an inflation target and a price level target? I don’t think so. So even if I accepted your weak ceteris paribus formulation, I would still say that it provides no useful guidance in how to formulate monetary policy.

  6. 6 Fed Up May 17, 2015 at 3:03 pm

    “In making this argument, as I have shown in an earlier post, Friedman displayed a basic misunderstanding of what constituted the gold standard as it was historically practiced, especially during its heyday from about 1880 to the outbreak of World War I, believing that the crucial characteristic of the gold standard was the limitation that it imposed on the quantity of money, when in fact the key characteristic of the gold standard is that it forces the value of money – regardless of its material content — to be equal to the value of a specified quantity of gold. (This misunderstanding – the focus on control of the quantity of money as the key task of monetary policy — led to Friedman’s policy instrumentalism – i.e., setting a policy rule in terms of the quantity of money.)”

    Let’s assume a 100% reserve requirement for gold. Why isn’t that true about limiting “money”? Assume 1 oz of gold = $1 of currency. The fixed part is key not necessarily the 1 oz to $1. If there are 1,000 oz of gold, there can only be $1,000 (stock of “money”). If the ounces of gold only grow by 1% to 4% per year, it will limit “money” growth to 1% to 4% per year.

    Also, it seems to me it is the stock of “money” that matters.

  7. 7 Frank Restly May 18, 2015 at 3:16 pm

    Fed Up,

    The U. S. went onto a gold standard in the year 1900 with this act:

    Prior to that the U. S. was on a bi-metallic (gold and silver standard).

    Fractional reserve banking coexisted with the gold standard. Even if the gold supply only grows at 3% per year, the effective money supply can grow at 5 times the gold growth rate (15% credit money growth) with a 20% reserve requirement.

    Realize that with fractional reserve lending it becomes impossible for a central bank to redeem all claims on gold simultaneously. And so the conversion rate between dollars and ounces of gold was of the singular direction variety – central bank would always give you $20.67 for each troy ounce of gold but would not necessarily reciprocate on the reverse trade.

  8. 8 Fed Up May 21, 2015 at 10:09 pm

    Frank, that is one reason I said assume a 100% reserve requirement.

  9. 9 LAL May 24, 2015 at 6:38 pm

    A much more interesting post than your last Taylor smear article…

    what if the sergeant is a spy for the enemy and deliberately picks the wrong 5 men. The effect of such a willful decision would seem to be the same as a poor judgment. It doesn’t seem to matter the interpretation of actions so long as the rule could protect us from both.

  10. 10 David Glasner June 1, 2015 at 9:11 am

    My apologies for being so tardy in replying:

    Fed Up, A 100% reserve requirement of 1 ounce of gold per dollar issued would not necessarily keep the value of a dollar equal to an ounce of gold. The gold standard operates by a convertibility requirement, not a reserve requirement. Convertibility could be maintained virtually no reserves. The idea that the gold standard is a system whereby currency is “backed” by gold is a basic misunderstanding of which Friedman regrettably subscribed to.

    LAL, I am glad you seem to like this post better than my previous one about Taylor. I have to say that I do not appreciate your suggestion that I have smeared Taylor. If I have written anything about Taylor that was untrue, please cite it explicitly. If you can’t do so, then please don’t say that I am smearing him. As for your question, discretion can be abused and rules can be broken. No system is perfect.

  11. 11 Fed Up June 6, 2015 at 10:58 pm

    “Fed Up, A 100% reserve requirement of 1 ounce of gold per dollar issued would not necessarily keep the value of a dollar equal to an ounce of gold.”

    I am not getting that one at all.

    Assume 1 oz of gold = $1 of currency (fixed exchange rate by the central bank). AND, assume a 100% reserve requirement for gold and currency. 1,000 oz of gold means there can be no more than $1,000 in currency.

    “The gold standard operates by a convertibility requirement, not a reserve requirement. Convertibility could be maintained virtually no reserves.”

    If I am reading that correctly, the fixed exchange rate may be in jeopardy of being broken. Let’s relax the above example by lowering the reserve requirement to 50%. Now 1,000 oz of gold means there could be $2,000 in currency. If something changes so that entities want to redeem more than $1,000 in currency for gold, the fixed exchange rate is going to have to give (some entity or entities are going to be defaulted on).

    Is there something I am missing?

  1. 1 David glasner on john taylor. Cluelessness about Strategy, Tactics and Discretion « Economics Info Trackback on May 6, 2015 at 1:02 pm
  2. 2 Krugman’s blog, 5/6/15 | Marion in Savannah Trackback on May 7, 2015 at 4:37 am
  3. 3 Vocidallestero » Krugman: La Banca Centrale Non Ha il Controllo della Quantità di Moneta Trackback on May 7, 2015 at 9:50 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

About Me

David Glasner
Washington, DC

I am an economist in the Washington DC area. My research and writing has been mostly on monetary economics and policy and the history of economics. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey's unduly neglected contributions to the attention of a wider audience.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 427 other followers

Follow Uneasy Money on

%d bloggers like this: