I’ve always thought that the debate over whether economics is a science is actually a debate about the meaning of the term ‘science.’  If science is when people build models to better understand the world around us, then economics is a science.

Some people distinguish between soft sciences like economics, and hard sciences like physics.  I don’t understand that distinction.  Are physicists very good at predict earthquakes, tornadoes, heat waves, etc?  Obviously not.  Some people claim that those events belong in geology or meteorology, but not physics.  My response is that physicists claim that their models explain those phenomena, so we have just as much right to expect them to predict tsunamis as we have to ask economists to predict recessions.  Indeed even more of a right, as economics has a sort of Heisenberg uncertainty principle, which says that (demand-side) recessions should not occur if they are expected.

Economics can predict some things very well and other things not so well.  The more complex the system, the less well we predict–just like physics.  I predict that bond prices will be unusually volatile between 8:30 and 8:35am on the first Friday of each month over the next few years.  But our most useful predictions are conditional forecasts.

Tyler Cowen has an interesting post on this question, but ends up with a perplexing assertion:

I conclude that economics is not yet a science.  Economics is most like a science when people do not care about the outcome of the argument.

Tyler is much better read than I am, but I was under the impression that scientists often became emotionally attached to their theories, and that it was hard to dislodge them.  Didn’t Max Planck say that science progresses one funeral at a time?

There is also a widespread view that we economists should try to become more scientific.  I’m not sure what that means.  If economics is not a science, I don’t think it should try to be one; just as I don’t think music, law, and accounting should try to become sciences.  We should try to become more useful, not more scientific.  I happen to think we are a science, BTW, as we do build models and use them to try to explain cause and effect in the world around us.  But if I am wrong and we aren’t a science, then I presume there is a good reason for us not being a science.

To summarize:  I see us as being a science.  Those who disagree with me simply define science differently.  It makes no difference whether we are or are not a science; all that matters is whether we are useful.  Law and music are not sciences, but they are very useful.  I hope this blog is useful.

I’m not happy about having to criticize Frederic Mishkin’s money textbook.  He was my teacher at Chicago and he seems like a great guy.  I’ve used his text for roughly 20 years and it’s a fine book.  Even worse, he was recently victimized by an unfair and misleading ambush interview.  But I must pursue The Truth wherever it takes me.

One of my favorite things about Mishkin’s text was that it presented aggregate demand curve in two ways.  At the beginning of chapter 22 it developed what is sometimes called the “monetarist” version of AD, which shows the curve as a fixed level of nominal GDP, i.e.  a rectangular hyperbola in P-Y space.  Only then does he present the so-called “Keynesian” version of AD, which is so hard to understand that I won’t even try to explain it.  And if I can’t understand it, I’m pretty sure our undergraduate students can’t either.  BTW, I have no idea why an AD curve shaped like a rectangular hyperbola is called “monetarist.”  It has nothing to do with whether V is constant or not (although Mishkin implies it does.)

Thus I was very disappointed to see Mishkin drop the monetarist AD curve from the new edition.  Textbook changes are usually made under the influence of criticism from professors at obscure community colleges, and perhaps that happened here as well.  I guess most professors prefer to teach a model that no undergraduate is likely to understand, rather than a simple and elegant framework that partitions macro into two parts; models that explain changes in nominal spending, and models that explain how nominal shocks get partitioned into output and inflation.  What could be simpler?

If I was a conspiracy buff, I would note that this change occurred right after the biggest fall in NGDP since 1938.  If one used the monetarist framework, it might lead students to ask uncomfortable questions about why the monetary policymakers allowed M*V to fall so sharply.  But I’m not a conspiracy buff.

Another thing I really liked about the 7th edition was the following question (on p. 368) about IOR:

10.  The Fed has discussed the possibility of paying interest on reserves.  If this occurred, what would happen to the level of e [the excess reserve ratio]?

I loved this question.  And when it actually happened, I couldn’t wait to show my students how Mishkin’s book predicted the dire consequences of the Fed’s October 2008 decision to adopt IOR.

So you can imagine how disappointed I was to find the question mysteriously deleted from the 8th edition.  If I was a conspiracy buff I’d wonder whether Mishkin was trying to hide something.  Surely students who did this question would be inclined to ask why the Fed did a highly contractionary policy in the midst of the biggest fall in AD since 1938 (that is if they still understood that AD=NGDP, which is doubtful.)

You might ask whether I am being too suspicious, after all, authors extensively revise texts with each new edition.  Times change, books need to reflect issues of current importance.  Actually, one of the dirty little secrets of the publishing world is that new editions are almost identical to old editions.  The publishers frequently revise editions so that they can sell new copies to students at $124 each, rather than have students buy old copies from previous students.  And of course far from being an obsolete question, the IOR question could hardly have been timelier.

Fortunately I’m not a conspiracy buff, so I’m willing to give Mishkin a pass.

A year ago I did a long post discussing how Mishkin’s text provided a template for my critique of the conventional wisdom circa October 2008.  I specifically cited 3 of the 4 key principles that Mishkin identified in his summary of the monetary policy transmission mechanism (pp. 610-11):

1.  It is dangerous always to associate the easing or the tightening of monetary policy with a fall or a rise in short-term nominal interest rates.

2.  Other asset prices besides those on short-term debt instruments contain important information about the stance of monetary policy because they are important elements in various monetary policy transmission mechanisms.

3.  Monetary policy can be highly effective in reviving a weak economy even if short term rates are already near zero.

I had thought that all economists accepted these propositions, as they are taught in the number one undergraduate money text.   And not just taught; they are the summation of the most important chapter in the text.   And they also happen to be true.  But I found out in late 2008 that very few economists accept these propositions.

I was anxious to get Mishkin’s new text, where he could take a sort of victory lap.  He could show how the Fed made a huge mistake allowing all sorts of asset prices to crash in late 2008, which signaled ultra-tight money.  But before I got the new edition, I read some articles where Mishkin seemed to be defending Bernanke’s moves.  I guess I shouldn’t have been so naive.  Mishkin and Bernanke are both center-right New Keynesians.  Both served on the Federal Reserve Board.  And of course Bernanke had also held similar views in the early 2000s, when he insisted that BOJ policy was far too tight, despite low rates.  So if Bernanke did a complete flip flop, why should I be surprised if Mishkin did as well?

Still, there was the question of how he would reconcile his views of the 2008 crisis with those three key principles of monetary policy.  I know what you are thinking—he dropped the key principles.  No, those are far too important to eliminate.  Did he refrain from discussing the crisis?  No, how could he do that?  Instead, he stated his view of the crisis just one page before the three principles that completely conflict with his view of the crisis.  That takes chutzpah!

Here’s what he says about the crisis on page 609:

With the advent of the subprime financial crisis in the summer of 2007, the Fed began a very aggressive easing of monetary policy.  The Fed dropped the fed funds rate from 5 1/4% to 0% over a fifteen-month period from September 2007 to December 2008.

Wait a minute; doesn’t he say just one page later than low rates don’t mean easy money—that you have to look at other asset prices?  Yes, but perhaps Mishkin didn’t know about all the other asset markets (TIPS, stocks, commodities, forex, commercial real estate, etc), which all started screaming that money was too tight in late 2008, as rates were gradually cut from 2% to 0%.

After discussing the crisis, Mishkin continues (p 610):

The decline in the stock market and housing prices also weakened the economy, because it lowered household wealth.  The decrease in household wealth led to restrained consumer spending and weaker investment, because of the resulting drop in Tobin’s q.

With all these channels operating, it is no surprise that despite the Fed’s aggressive lowering of the fed funds rate, the economy still took a bit [sic] hit.

So I guess he did know.  But perhaps there is nothing more the Fed could have done once rates hit zero?  Surely I can’t seriously claim that monetary policy can be highly effective once rates hit zero?  Go back and read Mishkin’s third principle.

If I was a conspiracy buff, I’d say that Mishkin followed almost every other famous economist in assuming that Fed policy was easy during late 2008, despite plunging stock and commodity prices, soaring real interest rates on 5-year TIPS, plunging inflation expectations, a soaring dollar, and plunging real estate prices.  And he assumed there was nothing the Fed could do about it because they had already cut rates to zero.

If I was a conspiracy buff, I’d wonder if he knew there was a contradiction, and erased any passages of the book that might alert students to the possibility that the Fed policy was actually tight (such as the IOR question) or that the sharp fall in M*V was the big problem in 2008–i.e. the monetarist view of AD.

If I was a conspiracy buff I’d even wonder if he was so nervous and distracted typing the last line I quoted that he misspelled ‘big’.  That he was nervously looking ahead to the very next section in his textbook; the Lessons for Monetary Policy.

Fortunately I’m not a conspiracy buff.  But now I understand why so many people believe Bush was behind 9/11, or LBJ was behind the Kennedy assassination, or FDR knew about Pearl Harbor before it happened, or the CIA overthrew Allende.  It really is a lot of fun being a conspiracy buff.  What a satisfying view of the world!  Everything has an understandable cause, all loose ends tied up with a nice Christmas bow.

PS.  Cowen and Tabarrok do AD the correct way.

PPS.  In this earlier post I argued that confused professors probably forced Mankiw to remove the one question that actually taught S&D correctly, which showed students that one should not expect consumers to buy less after a rise in the price.

PPPS.  BTW, I really do think Mishkin was treated unfairly in the interview.  Keep in mind that this was done by a director who used Barney Frank to explain what went wrong in the sub-prime crisis.

In recent years I have mostly taught Monetary Economics.  I spend precisely zero minutes covering IS/LM, and lots of time covering market responses to Fed announcements.  Based on some of the comments I received from my last post, I thought it might be interesting to examine one of my favorite Fed announcements, the surprise 1/2 rate cut of January 3, 2001.  This was the first rate cut of the 2001 recession, and reduced the fed funds target from 6.5% to 6.0%.  Here were some market reactions:

1.  The S&P 500 soared by 5.0%

2.  T-bill prices rose by 1/32  (yields fell)

3.  28 year TIPS prices fell by 16/32  (yields rose)

4.  28 year T-bond prices plunged by 76/32 (yields rose sharply)

I generally attribute the slight fall in T-bill yields to the liquidity effect, the rise in stock prices and TIPS yields to the income effect, and the sharp rise in T-bond yields to the income and expected inflation effects.  IS/LM says easy money lowers rates.  I say it is much more complicated than that.

I know I am in the minority, because here’s how the Wall Street Journal (1/4/01) described the market reaction:

Treasury Prices Plummet as Stock Prices Soar on Earlier than Expected Fed Interest Rate Cut

.   .   .  “Market participants were remembering history: that when the Fed’s on your side, stocks are the place to be,” said Mark McQueen, executive vice president of Sage Advisory Services.

Traders said the selling in bonds also reflected the fact that the market had been anticipating an easing of Fed policy soon and already had rallied strongly on the expectation.  When the Fed actually cut rates, people were, in bond market terms, “selling the fact.”

Have you noticed that when reporters don’t have a clue as to what is going on, they attribute it all to market irrationality?  No wonder there’s so much hostility to the efficient markets hypothesis.  Reporters are taught that easy money leads to lower interest rates.  When the markets behave in seemingly irrational ways, the natural reaction is to blame the EMH.  Readers of this blog know that easy money often leads to higher long term interest rates, so we don’t have to resort to market irrationality to explain these reactions to Fed announcements.

Some commenters noted that yesterday’s complex bond market reaction reflected changes in the proportions of maturities that the Fed was likely to purchase.  Maybe so, but in qualitative terms the reaction really wasn’t much different from January 2001, a policy move associated with virtually no bond purchases.  (During normal times the level of excess reserves is tiny and the Fed buys only an infinitesimal amount of bonds when it eases policy.)

Of course most Fed moves are widely expected, and elicit no market reaction.  For instance, the Fed did a long series of quarter point increases between 2004 and 2006, which the market took in stride.  The biggest reactions occur at key points in the business cycle, when a significant change in policy is contemplated.

The next key policy move occurred in September 2007, when the Fed did its first rate cut of this cycle.  Again, the cut was larger than expected (1/2 point), and once again short term rates fell and long term rates rose.  Stocks rose a couple percent after the announcement.  Asian stocks soared on the news. (Sound familiar?)

In December 2007, the fed funds futures market showed the cut would be either a quarter or a half point.  The actual 1/4 point cut caused a severe stock price decline.  This time the income effect stretched all the way down to three months, as yields fell from 3 months to 30 years.  And remember, these rate cuts were responding to a contractionary surprise, so the IS/LM model says yields should have risen.  The stock and bond markets turned out to be prescient.  A recession began at exactly that moment, and by January the Fed realized it had blown it.  It did two rates cuts totaling 125 basis points within 2 weeks, which validated the earlier T-bill market reaction in December.

This was enough to prop up NGDP for another 6 months.  But between July and December 2008 NGDP went into free fall, as the Fed became mesmerized by commodity inflation and failed to respond to abundant signs that NGDP growth was plummeting.  Then they became obsessed with the banking crisis.  Indeed the great collapse of July-Dec. 2008 was more than half over before the Fed started doing monetary policy, and the first move was not a rate cut but rather the interest on reserves program of October 2008—a highly contractionary move!

Lessons for the Fed?  Pay attention to markets that provide information about future NGDP growth, and don’t ever let NGDP expectations going 1, 2, and 3 years out collapse to a level far below trend.  And don’t assume that low rates mean easy money.

If I wrote a macro textbook, I would try to avoid any mention of interest rates or inflation.  The Fisher equation would use expected NGDP growth.  The AS/AD model would use hours worked as the real variable and NGDP as the nominal variable.  The transmission mechanism would not involve changes in interest rates, but rather the excess cash balance mechanism.  More cash would raise expected future NGDP.  This would raise the current price of stocks, commodities and real estate.  (As in Islamic economics, there’d be no interest rates.)  The higher asset prices would tend to raise current AD.  More nominal spending, when combined with sticky wages and prices, would boost output.

[Update 10/22/10:  This is why I shouldn’t do 4 posts in one day.  Commenters pointed out two flaws.  The Fisher equation uses interest rates.  And interest rates are important for inter-temporal decisions.  How about this:  A macroeconomics free of the concept of inflation, and a business cycle theory that did not use either inflation or interest rates.  Is that slightly less crazy?]

But obviously I’m the exception.  When rates hit zero and the Fed couldn’t move them anymore, I expected economists to shift over to some other mechanism; the money supply, CPI futures, exchange rates, etc.  Instead they started talking about how the Fed could promote a recovery by lowering long term rates.  (But if the policy is expected to work, wouldn’t it boost long term rates?)  Or they talked about how the Fed could reduce real interest rates by boosting inflation.  Some even argued that the Fed would have to boost inflation expectations to 6% in order to get a robust recovery, forgetting that this view directly conflicts with another key assumption of Keynesian macroeconomics—that the SRAS is very flat when unemployment is high.

Some of my commenters argued you couldn’t raise NGDP without first creating inflation expectations, which would lower the real rate of interest.  But that’s not necessary at all.  If you create higher NGDP growth expectations, then even if expected inflation doesn’t rise at all, the Wicksellian equilibrium real rate will rise and monetary policy will become more stimulative.  So the Fed doesn’t need to lower real interest rates in order to stimulate the economy.  Conversely, a policy such as interest on reserves might have had a devastating impact on aggregate demand, even if it had no impact on interest rates.

I’m genuinely confused.  When we explain why a big crop of apples makes NGDP in apple terms rise sharply, we don’t resort to convoluted explanations involving an interest rate transmission mechanism.  Why then, when there’s a big increase in the supply of money, do we think it will only affect nominal GDP in dollar terms via some sort of interest rate transmission mechanism?

This post was motivated by the following comment in a recent Paul Krugman post:

By the early 1980s it was already common knowledge among people I hung out with that the only way to get non-crazy macroeconomics published was to wrap sensible assumptions about output and employment in something else, something that involved rational expectations and intertemporal stuff and made the paper respectable. And yes, that was conscious knowledge, which shaped the kinds of papers we wrote. So you could do exchange rate models that actually had realistic assumptions about prices and employment, but put the focus on rational expectations in the currency market, so that people really didn’t notice. Or you could model optimal investment choices, with the underlying framework fairly Keynesian, but hidden in the background. And so on.

I can’t quite tell whether Krugman thinks “rational expectations and intertemporal stuff” are counterproductive, so the following isn’t really aimed at Krugman, but rather the many economists who I am quite certain do regret the influence of ratex and the efficient markets hypothesis.

The most important macro event of my lifetime was the precipitous drop in inflation and NGDP growth expectations during the late summer and early fall of 2008.  Because inflation had been running around 5% over the previous year, antique macro models based on adaptive expectations were completely useless during this period.  Indeed, I’m guessing that one of the reasons why the Fed was behind the curve in the fall of 2008 was precisely because their models were far too backward-looking.  During the financial crisis Fed policy got effectively tighter and tighter, and as a result there was a whole lotta “rational expectations and intertemporal stuff” going on.

Here’s what I find so ironic.  Everyone talks about how the profession became obsessed with ratex and the EMH after 1980, but from my perspective most economists still seem stuck in the adaptive expectations era.  If you really believed in ratex and the EMH, wouldn’t you be really, really interested in market forecasts of the policy goal variable?  I would be.  Yet instead of trying to infer market forecasts, they built elaborate structural models to try to forecast the goal variables.  In the 1980s when I tried to peddle my futures targeting approach, no one seemed interested.  I presented papers at the AEA meetings, the NY Fed, the Philly Fed, and everyone just yawned.  So from my perspective we face exactly the opposite problem; the profession doesn’t take ratex and the EMH seriously enough.  If the Fed really believed in ratex and efficient markets, they would have put the pedal to the metal in the infamous September 16, 2008 meeting.  Instead they yawned, and left rates unchanged at 2%.

PS.  Perhaps me and the other futures targeting proponents were just ahead of our time.  One famous macroeconomist endorsed futures targeting just recently.  A sign of things to come?

PPS.  Yes, the Fed did use the EMH as an excuse not to intervene in the housing bubble.  But that’s exactly where it was not appropriate.  The existence of FDIC and TBTF means the EMH provides no justification for not regulating bank risk-taking.