Thursday, June 14, 2012

Taylor's "First Principles"

I recently read John Taylor's "First Principles" This is a really good book in many ways.

It's very accessibly written. I am often asked for recommendations of easy-to-read books that illuminate modern macroeconomics. Since I spend most of my time reading papers full of equations, I don't often have a good answer. This book belongs high on the list.

This is no ordinary what's-wrong-with-the-world, five-step-plan-to-greatness book, of which we see so many these days. It's also not a generic why-free-markets-are-great book. We always need more of those, but this isn't one.

This book is fundamentally about rules vs. discretion, commitment vs. shooting from the hip, and more deeply about whether our economy and our society should be governed by rules, laws and institutions vs. trusting in the wisdom of men and women, given great power to run affairs as they see fit.

The preference for rules is one of the most important lessons of modern macroeconomics.

Macroeconomics should really be called intertemporal economics. Every important piece of analysis is about how people balance the present and the future. Save for the future or consume today? Invest in a new factory or not? Start a new business or go home and play golf?  And intertemporal decisions are all about expectations. If you're deciding whether to consume or invest, your expectations about how that investment will pay off are crucial.

That much is not controversial. For this reason, there is a lot of policy talk about "managing expectations," "giving confidence" and so on. Much new-Keynesian advice, now in vogue at the Fed, centers on announcements the Fed should make in order to guide our expectations.

The central insight on which John builds is this: rules, institutions, laws, and pre-commitments lead to much better outcomes when expectations matter so much to decisions.

John puts it better and more succinctly than I have: 
If people are forward-looking, and adjust their behavior to new circumstances, then economic policy works best when formulated as a rule. Government's adherence to known rules allows people to have a better sense of what is coming, and therefore to make more-informed decisions about long-range plans. (p. 23)
John continues, 
Setting out a sensible rule and sticking to it also helps policymakers resist interest-group pressure. Rather than having to consider the merits of every special-interest plea for more government support, a rule can set a standard that applies to all cases and limits the role of government broadly. 
This is a second bit of wisdom. A rule "we don't have dessert on weekdays" leads to a better dinnertime conversation with your kids than if every night is separately negotiated. 

Macroeconomics is behind law in this regard. For centuries, we've understood that giving wide discretion to legal officials is a bad idea. Judges should not be empowered to "do what you think best at the time." Giving such power might seem attractive -- after all, an unconstrained judge can surely always find a better solution to a given problem. But our legal system understands the horrible incentives for prior behavior if the judge is unconstrained ex-post.  We don't even give our legislature complete freedom; we constrain it with  constitution. Taylor rightly connects the more technical macroeconomic literature on rules and pre-commitment vs. discretion to this larger social and legal wisdom. (See the nice Hayek and Friedman quotes on p. 22)

Somehow, that lesson is lost on much policy-oriented macro, and increasingly on financial regulation.

In macroeconomics, it is largely a result of Keynesian thinking. Though Keynes wrote about expectations, the ISLM models used by his followers pretty much live out of time, with today's income driving today's consumption and so on. The resulting policy analysis is not really intertemporal at all. It emphasizes "what do we do now?" over "how should policy systematically react to recessions, given that people will learn to anticipate such policy and may therefore undermine its effects?"

Our new financial regulation basically just gives unlimited discretion to the Financial Stability Council, meaning mostly the Fed, to do whatever it thinks right.  The hope that a huge fire department will lead people to buy their own fire extinguishers, be careful about playing with matches, and not try to bribe the firefighters to come to their house first, seems pretty hopeless to me.

Of course, there is a reason people don't want to follow rules. The discretion to do anything you want is always  more powerful ex-post. The problem is, great power ex-post leads to bad economic behavior ex-ante, so you end up worse after the fact.  By precommitting to actions ex-post, you end up in a better situation overall, but once the precommitment has its intended calming effect, and ignoring how today's action will affect tomorrow's expectations, you could always do better just this once by deviating from the rule.  Please, dad, just this once? John doesn't really describe this tension, but I think it's important to understand why it's so hard to live by rules.

The rules lesson is also one of the hardest to communicate to a non-economist audience. And to a lot of economists too, I might add!  The natural inclination to always ask "What should we do now?" is nearly impossible to resist. And when asked, say, "should the Fed raise or lower rates today?" if you answer, "The Fed should follow x rule," you (I) quickly see eyes glaze over, and people start to play with their cell phones. "Yes, that's all very nice, but will you please tell us what should the Fed do now?" "Well, it should follow a rule...."

So this book is about the value of rules, institutions, and law in macroeconomics.

I highly recommend the chapter "Who gets us in and out of these messes."  Many of my students are unaware of basic macroeconomic policy history. John's uniting theme of the rise of activist policy in the 60s and 70s, the return to rules-based, long-run oriented policy in the 80s and 90s, and the return to activism since then is well told.

I also like it because it's decidedly non-partisan. Nixon's wage and price controls and Geroge W. Bush's stimulus checks come in for harsh criticism, and Taylor praises much Clinton-era policy as he does Reagans'.

The book really comes alive, of course, in discussing monetary policy. Taylor is justly famous for the "Taylor rule" advocating that the Fed's interest rate policy should be fairly mechanically related to inflation and output.

In "More Focus" (p. 123) John laments the Fed's multiple goals. By trying to manage both inflation and unemployment, the Fed tends to veer too much in one of the other direction unpredictably. And now that the Fed worries about the allocation of credit to particular markets, the financial health of big banks, and a host of other concerns, things will just get worse. So, says John  (p. 125)
The first step toward a more consistent policy would be to remove the dual mandate [inflation and unemployment] and bring the Fed's focus to a single goal. That goal should be price stability. 
I couldn't agree more. But then John becomes much more middle-of-the-road than the stirring rules, laws, and institutions vs. discretion preamble and this statement would suggest. You might conclude that  if the Fed's job is to ensure "price stability," it should just keep the CPI as close to a fixed level as possible, and that this mandate should be written in stone somewhere.  

John only asks that the Federal Reserve act be rewritten from its current "promote effectively the goals of maximum employment, stable prices, and moderate long-term interest rates" to "promote effectively long-run price stability within a clear framework of overall economic stability."  (He will have to add "financial stability" too, as most of the Fed is now a financial regulator.)  I see lots of room for short-run discretionary action in that mandate.

He follows, as he should (it's his book!) with a plea for the Taylor rule
..under a Taylor rule, the Fed, or any other central bank, is supposed to change its interest rate in response to both inflation and GDP. Specifically, the rule says that the Fed should set the interest rate equal to 1 1/2 times the inflation rate plus 1/2 ties the percentage amount by which the GDP differs from its long-run growth path, plus 1. 
This still sounds a lot like dual mandate doesn't it? Responding to GDP "helps mitigate the recession." And there is a lot of discretionary wiggle room in defining that "long-run growth path." Orphanides' analysis of the Fed in the 1970s suggests that they weren't feckless stimulators, instead they just didn't know that the long run growth path had trended down so badly. The same debate really rages today. Doves may rightly say "we're following the rule, it's just that GDP is further below trend than you think." 

The Taylor rule was originally an empirical description of Fed actions in the 1980s, a description of how the Fed acted to implement its dual mandate. It only slowly became a normative description of what the Fed should do, in the context of Keynesian and New-Keynesian models that posit a strong stabilization role for the Fed by exploiting a Phillips curve. The Taylor rule actually stands quite a bit to the left of the "inflation targeting" tradition that says central banks should only respond to inflation, ditching the whole GDP response -- because, in John's words (p. 127)
Some Federal Reserve officials worry that a focus on the goal of price stability would lead to more unemployment. But history shows just the opposite.
John answers that the  "dual response" really is a "single mandate." It is a a worthy effort, but one I find strained. The reason for the GDP response is, explicitly in the models, to accomplish a tradeoff between inflation and output volatility.

So, while I appreciate John's robust defense of the Taylor rule against the the Keynesian, activist, dual-mandate crowd, I would also have appreciated his defense of the GDP response against the views of inflation-targeters, price-level targeters, or even advocates of gold or commodity standards. Why not view the Taylor rule's GDP response as a transitionary arrangement on the way to an inflation or price level target, which seems to follow his principles better?

Then, there is the question how to bind the Fed to this rule. Remember, ex-post discretion is always tempting. (Puhleeeze dad, can't we have dessert tonight?) "Writing a policy rule into law" (p. 129) sounds promising.  But though having proposed a quantitative rule, John backs away from of the obvious idea to legislate it. (p. 132)
The most straightforward way to legislate a rule for monetary policy would be .. by reinstating the reporting and accountability requirements that were removed in 2000. ...This proposal does not require that the Fed choose any particular rule for the interest rate, only that it establish some rule and report what the rule is. But if the Federal Reserve deviates from its chosen strategy, the chairman of the Fed must provide a written explanation and answer questions at a public congressional hearing. So while the proposal limits discretion it does not eliminate discretion
Would this work?  The Fed chair regularly reports to Congress now, and explains its actions almost this way, something like:  "Yes, normally we'd be raising rates, but there's the banks, and headwinds from Europe, and unusually high unemployment and so on and so forth."  The Fed notoriously didn't let the  money growth targets get in its way. John writes persuasively (p. 133 ff) that such requirements would have made a difference. Read and decide for yourself.  I suspect Ron Paul would want  a constitutional amendment setting the conversion rate of dollars to gold.

John's capsule of the Fed's extraordinary actions in the financial crisis starting p. 136 is really worth digging out on their own:
The Fed's on-again off-again bailout measures were thus an integral part of a generally unpredictable and confusing government response to the crisis, which, in my view, led to panic.
But, I'm less persuaded that more reporting would have made much difference.  Faced with horrible situations and the ability to act with ex-post discretion, the Fed always will use that discretion. Pretty much everybody thinks the Fed will bail out large financial institutions that get in trouble, no matter what the Fed says, because it can. To me, the lesson of Lehman is that only lack of legal authority to act will prevent that action -- and credibly communicate to markets not to count on the bailout.

But I am being too critical. My intellectual habits are to find the purest simplest answer, ignore what's politically feasible, write it down, and prepare to be ignored. John's are to find a sensible small step that will likely improve matters substantially, and advocate that, with a strong chance of moving the current policy debate. His proposals fill that role admirably.

The final three chapters, "Ending Crony Capitalism as We Know It," "Improving Lives While Spiking the Entitlement Explosion" and "Rebuilding American Economic Leadership" are wonderful. Try to get your liberal friends to read them.

"Crony capitalism" properly stresses the nature rather than amount of regulation. We've given regulators far too much discretionary power, and this discretion is what breeds crony capitalism or worse, outright corruption.  Addressing the trope that the crisis came from "not enough" regulation, (p. 146)
..The government did not need more power or more discretion to regulate more markets or more firms in the wake of the crisis. It already had plenty of power before then. Indeed, it was this very power and discretion that led inexorably to the favoritism, to the bending of rules, to the reckless risk-taking and, yes, to the bailouts. Government bureaucrats hose which existing regulations to enforce and which ones to bend, and they [my emphasis] decided who was bailed out and who wasn't. ..This is textbook crony capitalism: the power of government and the rule of men -- rather than the power of the market and the rule of law -- to decide who will benefit and who will lose
More specifically, (p. 154)
The New York Fed had the power to stop the questionable lending and trading decisions of Citigroup and others. With hundreds of regulators on the premises of such large banks it also should have had the information to do so. The SEC could have insisted on reasonable liquidity rules to prevent investment banks from relying so much on short-term funds to finance long-term investments....
It has a great capsule of why Dodd-Frank is doomed to produce more crony capitalism.

The "entitlement explosion" chapter starts exactly where economists should start -- which is news to most people -- incentives, or rather the horrible disincentives that well-meaning programs unintentionally provide and lead to their predictable failure. Dear liberal friends: it's not about who cares more. It's about what the programs will actually do once people react to their incentives.
Entitlement programs also create powerful disincentive effects... The health care subsidy in the 2010 act declines as a family earns more income and then is cut to zero when 400 percent of the poverty line is hit. This creates a situation where if you work more, you earn less. Consider a family earning $80,000 that gets a health care subsidy from the government of $16,000 under the 2010 health care law, bringing their total income to $96,100.  Now suppose the husband or wife decides to work more. If they increase their income from work by $14,000, bringing their work earnings to $94,000, then their health care subsidy drops to zero. So they get less income by working more, and that's a big disincentive for the economy to grow. (p. 172)
 Dependency is not cultural or psychological. It's just incentives. The rest of the chapter summarizes simple common-sense and (relative to mine!) middle-of-the road solutions in a concise way.

And don't forget "Rebuilding American Economic Leadership." If America enters a few decades of Eurosclerosis, anemic growth, high unemployment, low innovation, large dependency, unsustainable entitlements, crony capitalism, politicized discretionary regulation,  and ultimately a European debt crisis, the ramifications are too ugly to think about.

So.... a review almost longer than the book. But a useful book to read and recommend, especially because it is clear, accessible, measured, and concise.


  1. Benanke has managed to convince me that economists are not to be trusted, and that we must return to the gold standard.

  2. '"Crony capitalism" properly stresses the nature rather than amount of regulation.'

    When people say "more" regulation, in practice they mean "more complicated" regulation.

  3. David Glasner says John Taylor really doesn't understand Hayek. Greg Ransom would say that about anybody, but Glasner isn't a similarly fanatical Hayek devotee (for instance, says Hayek's pronouncements on the Bank of France during the Great Contraction are a terrible black mark).

  4. This is a really comprehensive review of Taylor's book which is certainly on my radar. I have read His "Getting off track" book - very concise and to the point.

  5. Sounds like a good book, I need to check it out. Got some time for summer reading now.

  6. If Taylor could just graduate to Market Monetarism....

  7. Why do moral scientists always come out of the mortal fabric of humanity being named John Taylor by their ignorant parents. This is amazing. How many do we have now.


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.