The Region

JUNE 2010 26

In a discipline that celebrates specialization, Robert Hall is a Renaissance man.And economics is far the richer for it.

The Stanford economist’s extensive publications over four decades—books, blogs, articles and lectures—provide ready evidence of wide-rangingexpertise. (The interview below hardly scratches the surface.) As a laboreconomist, Hall has produced some of the field’s most influential models oflabor market dynamics and essential articles on labor supply, demand andwages. A scholar of fiscal policy, he built the intellectual foundationfor the 1986 tax reform bill as well as recent consumption tax proposals.

His work in financial theory, consumer and corporate incentives,and government policy illuminates regulatory issues currently underdebate in Washington. Innovative analysis of stock market valuation by Halldemonstrated the importance of intangible capital. His studies of entrepreneurialincentives (with his wife, economist Susan Woodward) and antitrust theoryare pathbreaking.

Hall’s research on trading through electronic markets—“digital dealing”is his term—provided one of the first lucid explanations of the economics ofthen-new Internet phenomena such as eBay. Hall’s analytical gifts also havegenerated important insights on monetary theory and optimal monetarysystems. He has done invaluable work as well in growth theory, determinantsof productivity, spending on health and economic geography.

His erudition has range, depth and quality that few economists canmatch. And the profession has recognized this with honors including theRichard T. Ely lecture in 2001, presidency of the American EconomicAssociation in 2010, and fellowship in the American Academy of Arts andSciences, Econometric Society and National Academy of Sciences.

Hall’s public profile, however, is largely confined to the sphere ofbusiness cycles, in which he also has unquestioned expertise. That narrow“fame” is due to his chairmanship, for over 30 years, of the committee thatdetermines when U.S. recessions officially begin and end. It is a painstakingand largely thankless task. Pundits and policymakers clamor for the committee’sannouncements, but inevitably second-guess the decisions made. Committedto the integrity of process and result, Hall has never bent to pressure, manifestingtime consistency that monetary policymakers can only envy.

Robert E.Hall

Photography by Peter Tenzer

THOUGHTS ON U.S.MONETARY POLICY

Region: Perhaps we could start withmonetary policy. What is your broadview of the Fed’s efforts over the pastfew years to stem the crisis using uncon-ventional monetary policy and strate-gies?

Hall: First of all, I believe you shouldthink of the Fed as simply part of thefederal government when it comes tothe financial side of its interventions. Ifyou look at how the federal governmentresponded initially, it was the Treasurythat was providing the funds. Of course,TARP [Troubled Asset Relief Program]was there using the taxpayers’ moneywithout involvement of the Fed. Also,early in the crisis Treasury depositedhundreds of billions of dollars at theFed, which the Fed then used to buyassets. So there the Fed was just an agentof the Treasury. It was as if the Treasurytook its funds to a broker.Eventually, the Treasury was imped-

ed from doing that by the federal debtlimit. But the debt limit doesn’t apply tofunds borrowed by the Fed, so it thenstarted borrowing large amounts frombanks by issuing reserves. That is whatcaused all the confusion about thinkingthis was somehow part of conventionalmonetary policy.I would distinguish between conven-

tional monetary policy which sets theinterest rate and this kind of financialintervention of buying what appear tobe undervalued private securities.Issuing what appear to be overvaluedpublic securities and trading them forundervalued private securities, at leastunder some conditions and some mod-els, is the right thing to do. In my mind,it doesn’t make a big difference whetherit’s done by the Federal Reserve, theTreasury or some other federal agency.

Region: And what are your thoughts onthe best course for a Fed exit strategy?

Hall: That again gets at this confusion.Traditionally, reserves at the Fed pay

zero interest in the United States, so innormal times with positive marketinterest rates, banks try to unloadreserves; when they do so, they expandthe economy. That does not happenwhen interest rates in the market arezero because there’s no incentive forbanks to unload reserves. They can’tgain by getting something off their bal-ance sheet if what they buy doesn’t yieldany more. And during the crisis, therewas no differential, nothing to be gainedby unloading reserves.As the differential reestablishes,

which the markets think is going to hap-pen in the next year or so, then thatissue comes up. It would be highlyexpansionary and ultimately inflation-ary if market interest rates began to riseabove zero and the Fed didn’t do some-thing to either reduce the volume ofreserves or increase the demand forreserves.So the Fed has two tools, and

Chairman Bernanke has been very clearon this point. He’s given a couple of

excellent speeches that have describedthis fully, so it shouldn’t be an issue, andI think more or less it’s not anymore. TheFed can either leave the reserves outthere but make them attractive to banksby paying interest on them, or it canwithdraw them by selling the correspon-ding assets they’re invested in. Sellingassets will be timely because those invest-ments will have recovered to their propervalues; the Fed can sell them and use thefunds to retire the reserves.So, again, there are two branches to

the exit strategy: There’s paying intereston reserves, and there’s reducingreserves back to more normal levels.They’re both completely safe, so it’s anonissue. The Fed itself is just not a dan-ger. It is run by people who know exact-ly what to do. And we have 100 percentconfidence they will do it. It’s not some-thing I worry about.

FINANCIAL FRICTIONS

Region: That’s reassuring, but I believeyou doworry about financial frictions…

Hall: I do, I do very much.

Region: Your recent paper on gaps, or“wedges,” between the cost of andreturns to borrowing and lending inbusiness credit markets and homeownerloan markets argues that such frictionsare a major force in business cycles.Would you elaborate on what you

mean by that and tell us what the policyimplications might be?

Hall: There’s a picture that would helptell the story. It’s completely compelling.This graph shows what’s happened dur-ing the crisis to the interest rates facedby private decision makers: householdsand businesses. There’s been no system-atic decline in those interest rates, espe-cially those that control home building,purchases of cars and other consumerdurables, and business investment. Soalthough government interest rates forclaims like Treasury notes fell quite a bitduring the crisis, the same is not true forprivate interest rates.

The Region

28JUNE 2010

There are two branches to the exitstrategy: There’s paying interest onreserves, and there’s reducing reservesback to more normal levels. They’re bothcompletely safe, so it’s a nonissue.The Fed itself is just not a danger. It isrun by people who know exactlywhat to do.

Between those rates is some kind offriction, and what this means is thateven though the Fed has driven theinterest rate that it controls to zero, ithasn’t had that much effect on reducingborrowing costs to individuals and busi-nesses. The result is it hasn’t transmittedthe stimulus to where stimulus is need-ed, namely, private spending.The government sector—federal,

state and local—has been completelyunable to crank up its own purchases ofgoods; the federal government has stim-ulated [spending] slightly but notenough to offset the decline that’soccurred at state and local governments.

Region: Yes, I’d like to ask you about thatlater.

Hall: So to get spending stimulated youneed to provide incentive for privatedecision makers to reverse the adverseeffects that the crisis has had by deliver-ing lower interest rates. So far, that’s justnot happened. The only interest ratethat has declined by a meaningfulamount is the conventional mortgagerate. But if you look at BAA bonds orauto loans or just across the board—there are half a dozen rates in this pic-ture—they just haven’t declined. Sothere hasn’t been a stimulus to spend-ing.

The mechanism we describe in ourtextbooks about how expansionary pol-icy can take over by lowering interestrates and cure the recession is just notoperating, and that seems to be verycentral to the reason that the crisis hasresulted in an extended period of slack.

Region: So to incorporate that in amodel seems quite important.

Hall: Yes, and many, many macroecono-mists have turned their attention to that.I’ve been following the literature andbeen a discussant at many conferencesof other people’s work on this. In fact,the Fed is giving a conference at the endof next week, and I’ll be presenting mypaper on frictions.

Region: Your model is able, I think, toexplain a fair amount of the currentbusiness cycle by incorporating thosefrictions.

Hall: I mainly look at, as kind of athought experiment, how much of adecline in activity occurs when thatkind of a friction develops. When pri-vate borrowing rates rise and publicborrowing rates fall, the differencebetween them is the amount of friction.I show that that’s a potent source oftrouble. I haven’t tried to align it with

history prior to the current crisis. That’san interesting question, but data on his-torical events aren’t always so easy, sothat lies ahead.

Region: And the policy implications?What can and should be done to reducefrictions?

Hall: Good question! Well, it does pointin the direction of focusing on thingslike lower rates for corporate bonds,BAA corporate bonds. They appear tobe undervalued private assets, althoughthat’s not been one of the types of assetsthat policy has seen as appropriate tobuy or to help private organizations tobuy. That would be one way to turn.

The Region

29 JUNE 2010

Even though the Fed has driven theinterest rate that it controls to zero, ithasn’t had that much effect on reducingborrowing costs to individuals andbusinesses. ... When private borrowingrates rise and public borrowing rates fall,the difference between them is the amountof friction. ... That’s a potent source oftrouble.

Most of the undervalued assets thatthe Fed has bought have been mortgagerelated. ... There would be a case forexpanding that type of policy to otherseemingly undervalued instruments.

We’ve concentrated on doing that inmortgage-related assets. You can see inthe picture that it’s had some effect.Most of the undervalued assets that theFed has bought have been mortgagerelated. It’s been kind of an obsessionwith trying to solve these problems asthey arise in home building, but homebuilding is only part of the story. Thecollapse in other types of investmentspending has been equally large. Therewould be a case for expanding that typeof policy to other seemingly underval-ued instruments.That would presumably result in the

same pattern you’ve seen in mortgages.That policy has been successful—differ-entially successful in depressing mort-gage rates as opposed to bond rates orother areas.

EQUITY DEPLETION

Region: Let me ask you about a paperyou wrote in December 2008, on equitydepletion, defined as the “withdrawal ofequity from firms with guaranteeddebt.” We’re all well aware of govern-ment bailouts, and implicit or explicitguarantees of financial institutions…

Hall: That paper was actually reprintedin a book that just came out, Forward-Looking Decision Making [PrincetonUniversity Press, 2010]. It’s the lastchapter in this book, which is a compi-lation of the Gorman lectures I gave atUniversity College London in October2008.

Region: You had a wonderfully provoca-tive statement in it. You declare thatequity depletion “appears to be anunlimited opportunity to steal from thegovernment.”Could you tell us what you mean by

that? Why does equity depletion occur,and how does it constitute an opportu-nity to steal?

Hall: George Akerlof and Paul Romerwrote a paper published in 1993 in theBrookings Papers that described whatthey called “looting.” The particular

form that looting took was through theownership of a savings and loan; thiswas a feature of the savings-and-loancrisis of the late 1980s.As a “looter,” you would use the sav-

ings and loan to attract deposits, pay thedeposits as cash to yourself and thendeclare bankruptcy. Akerlof and Romerdescribed a number of clever ways ofdoing that to escape the attention of laxregulators, and that’s the type of thingyou see in many settings.One of the big problems encountered

recently is that institutions that havebecome very undercapitalized were stilldepleting their equity by paying divi-dends. The government has had to pushvery, very hard to get these financialinstitutions to stop paying dividends.Dividends are exactly equity depletion.With a government guarantee, it’s exactlywhat there’s incentive to do—asdescribed in that paper.On the other hand, it seems we’ve

been much more successful currentlythan we were in what Akerlof andRomer described as far as preventingthe most extreme forms of this conduct.It’s a danger whenever you have guar-

anteed financial institutions that havegotten into a very low capital situation.They’ve suffered asset value declines,they’ve become extremely leveraged andthey have this very asymmetric payoff tothe owner: If they go under, it’s the gov-ernment’s problem; if they recover, it’sthe owner’s benefit. That asymmetry,which is the so-called moral hazardproblem, is just a huge issue.And yet, while we have a lot of insti-

tutions in that setting today, we don’t seemany of them doing things that Akerlofand Romer described, such as payingthemselves very large dividends. It’sbeen difficult to get them to cut the div-idends, but they have not paid out verylarge dividends or concealed dividends.So it looks like we’ve been somewhat

successful in preventing the worst kindof stealing, but the asymmetry is stillpotentially a big issue. There are way toomany bank failures that should not haveoccurred and especially should not havecost the taxpayers as much as they did.

Region: Your thoughts about what meas-ures can be taken to curb this moralhazard?

Hall: The most important thing is to besure that financial institutions that areguaranteed by the government havelarge amounts of capital so that the dan-ger of them spending the taxpayers’money rather than their own money isvery small. That’s a principle that’s beendeeply embedded in our regulations fora long time.But I pointed out in this chapter the

principle of so-called prompt correctiveaction, which says if capital goes belowthis mandated level, which is typicallyaround 8 percent, then something has tobe done right away before all theremaining capital gets depleted.We just have not been successful at

doing that. We have principles of regula-

The Region

30JUNE 2010

It’s a danger whenever you haveguaranteed financial institutions thathave gotten into a very low capitalsituation. They’ve suffered asset valuedeclines, they’ve become extremelyleveraged and they have this veryasymmetric payoff to the owner: If theygo under, it’s the government’s problem;if they recover, it’s the owner’s benefit.That asymmetry, which is the so-calledmoral hazard problem, is just a huge issue.

tion that allow the regulators to say thata bank is well capitalized even thoughthe markets know that it’s not. Bankshave been declared to be well capitalizedeven when the market value of theirdebt and the market value of their equi-ty have declined to very low levels.Regulators seem to ignore something

that everyone in the market seems toknow, which is that they’re shaky. Thereseems to be a lack of willingness to payattention to all the signals that a regulatorshould pay attention to. All they do islook at certain accounting records, whichdon’t reflect what people know.It’s not easy though. There’s been a

large amount of discussion of this topicamong very knowledgeable financialeconomists. My colleague DarrellDuffie here at Stanford has been a par-ticular leader. There’s a group called theSquam Lake Working Group, of whichhe’s a member, that has been advocatingideas like, as a backstop, having long-term debt be convertible to equity. Thatis what happens in a bankruptcy, butunder this strategy it would happenwithout a bankruptcy. It would happenautomatically with certain contingen-cies and would solve the problem in avery nice way. It would potentiallyincrease the borrowing cost, but itwould properly get the incentives right.A lot of people look to the example of

Citibank. Citibank’s long-term debt hasbeen selling at a considerable discount,which is a sign that the market knowsthat there’s an issue. So instead of doingwhat we have done, which is give guar-antees of short-term debt with govern-ment investments, the alternative thatthe Squam Lake people are thinking of,and I’ve been thinking of too, is to some-how convert Citibank’s long-term debtinto equity, which is the same thing thatthe market is in effect doing. That wouldeliminate the danger then that the bankcouldn’t meet its obligations, in a waythat is less burdensome to the taxpayer.In retrospect, what we did was to save

the economy from a tremendous trainwreck. But we didn’t do it in a way thatwas as cheap for the taxpayer as it couldhave been. And, of course, there have

been many examples discussed of this.This is all in retrospect. And I cer-

tainly don’t criticize the people whowere doing it at the time, especiallyChairman Bernanke. But looking for-ward now to the next time this happens,convertible debt would be a huge stepforward. If people at the Treasury couldhave just pushed a button to convert thedebt, without needing a new law, theywould have done it in a second. There’sno doubt about that. They just didn’thave that power.So we need to give regulators that

power through some sort of sensiblesecurity design. Regulators could dothat, and financial institutions wouldn’tsee it as terribly burdensome because themarket would know that the probabilityof this kind of thing happening again ispretty low. And when it does happenagain, which will be sometime in thenext century, that button would be thereto press, and we wouldn’t have the chaosthat we had in September of 2008.

GOVERNMENT SPENDINGAND GDP

Region: You mentioned earlier the diffi-culty of stimulating the economy, andI’d like to discuss your work on govern-ment multipliers. The federal govern-ment’s stimulus package has been atopic of heated debate among econo-mists, in terms of how much stimulusit’s truly provided and whether more isneeded. In a recent paper, you analyzebasically what happens to GDP whengovernment purchases goods and serv-ices.Would you give us your rough esti-

mate of the size of the multiplier in thecurrent era of very low interest rates,and share your sense of the impact ofthe current stimulus package?

Hall: The first thing to say, just lookingat the big picture, is that when the ideaof a stimulus through federal purchasesprogram came up in the current crisis,the thinking was, “That’s feasible. Wecan increase purchases.” And then thequestion was how much would it raise

GDP. There was a vigorous debate,around here anyhow, on this multiplierquestion.The discussion has shifted now

because the premise was that we wouldbe able to raise government purchases.But, in fact, government purchases havenot increased.In part that’s because it’s very difficult

and time-consuming to actually get thegovernment to buy more stuff. This hasbeen a critique of fiscal policy as long asI’ve been an economist, this notion thatit takes so long to get spending up thattypically the spending rises only afterthe recovery has occurred, and it comesat completely the wrong time.

Region:We searched in vain for “shovel-ready projects.”

Hall: Yes, “shovel-ready” turned out notto be. But the other fact is that there’sbeen a small increase in federal govern-ment purchases, but it’s been more than

The Region

31 JUNE 2010

It takes so long to get [federal government]spending up that typically the spendingrises only after the recovery has occurred,and it comes at completely the wrongtime. ... But the other fact is that there’sbeen a small increase in federal govern-ment purchases, but it’s been more thanoffset by declines in state and localgovernment purchases.

offset by declines in state and local gov-ernment purchases.The stimulus bill recognized that that

was a danger. We have had thesetremendously pinched state and localgovernments. A lot of them have justhad no choice when their tax revenuedeclined but to reduce spending.In spite of recognizing that potential

when the stimulus program wasdesigned, still the net effect of the crisisand the policy response was for govern-ment purchases to decline, not to rise.But by very small amounts. Basically,nothing happened to government pur-chases. And that was in an environmentin which everybody—and certainlyCongress was enthusiastic about it—waswilling to go for a program with higherpurchases. But no matter how hard theytried to turn the knob, it just wouldn’t govery far.

Region: So ARRA [American Recoveryand Reinvestment Act of 2009] was fornaught?

Hall: First of all, you have to take itapart, as I do in that paper, and ask howmuch of it went directly into govern-ment purchases, which is fairly small, orwould stimulate state and local purchas-es, which was also fairly small.A lot of it was providing income sup-

plements, and there you get into thequestion of whether the people receiv-ing the supplements increased theirspending or not. That’s a whole otherissue; I’m not commenting on that issue.That’s a very difficult question toanswer.To go on to the other part of your

question, had there been an increase ingovernment purchases that was success-fully achieved, how much would thathave increased GDP? The answer I gotwas around a factor of 1.7, which is atthe high end of the range of what mosteconomists were talking about.I only reached that by thinking very

carefully and reading a lot of recentcommentary on this question of theimplications of having a zero fed fundsrate. That turns out to be very impor-

tant. Others have found that to be true.So I think that the people who looked

at the evidence of what the multiplier isin normal times and said it’s maybe 0.8or 1.0 (which I would agree with) kindof missed the point. There was a lot of, Ithink, inappropriate criticism.Valerie Ramey, in contrast, has

focused not on the immediate policyquestion but raised the scientific ques-tion about the long-run multiplier. Hernumbers are ones that I respect and agreewith. They’re more in the 0.9 range.But on the issue of multipliers during

periods of zero interest rates, because wedidn’t have any changes in governmentpurchases during this one time whenwe’ve reached the zero interest point, wedon’t have any good empirical evidence.What we need is a time when interestrates are zero and there’s a big increasein government purchases. That just hasn’thappened.So we have no way to know through

pure practice; we have to use models.The models are very clear that it makesa big difference when we’re at the zerointerest rate limit. The normal configu-ration is that you get this fiscal expan-sion—the government buys more, butthat triggers sort of an automaticresponse from monetary policy to leanagainst it. If you shut that down by hav-ing interest rates stay at zero, you’ll get abigger effect. That’s what this literaturesays and it’s quite a big difference.

TAX POLICY

Region: Of course, this raises the issue oftaxes, of needing to pay for deficit spend-ing. And I notice the Time magazinecover above your desk about the flat tax.

Hall: From long ago!

Region: Yes, exactly. Your work withAlvin Rabushka on the flat tax was ahuge sensation in the early 1980s, asrepresented by making the cover ofTime.

Hall: That’s right. It’s one thing to getyour face on the cover of Time; it’s quite

another to get your idea on it! Forgetwhat’s-his-name’s face!

Region:And I think it can be argued thatthat helped pave the way toward the1986 Tax Reform Act.

Hall: We like to think so. I’ll accept that.

Region: Twenty-five years later you reis-sued the book, updated of course, andcontinue to advocate it as the “most fair,efficient, simple and workable plan onthe table.”Given its clear merits and strong

advocates, why do you think it’s gainedrelatively little traction in the UnitedStates?

Hall: One important thing to under-stand is that contrary to some people’simpressions, it’s not gone very far in therest of the world either.

Region: Not in central and easternEurope? Mexico, perhaps?

Hall: Yes, but if you look at their overalltax structure, it’s not what we have inmind. Their rates are high becausethey’ve adopted income tax systems thatwork like a flat tax, but they’re on top ofa very high value-added tax. So thecombination doesn’t achieve the lowrates that we were hoping for.In the U.S., there’s been a lot of back-

sliding. It looks like there’s going to bemore and more. The state of California,for example, has a couple of times addedsurcharges for very high incomes. Thereseems to be a belief that it’s a great idea,that we can get all the revenue we needby taxing high incomes, without regardto the problems that those tax rates cre-ate, especially in the longer run. That’sone of the things we talk about in ourbook. There’s more to the logic of lowmarginal tax rates than just the questionof who pays the tax.But another factor I would emphasize

is that since 1981 when we first promot-ed that plan, there’s been a dramaticwidening of the income distribution inthe U.S. That means that the idea of the

The Region

32JUNE 2010

poor paying the same tax rate just seemsless viable than it was when the incomedistribution was tighter.The division between a small num-

ber of winners in the modern economy,mostly businessmen and lawyers, asopposed to most other people, hasgrown significantly.In “others,” I include doctors, by the

way. One of the amazing things thatdoesn’t get much attention these days isthe widening division between doctorsand lawyers. It used to be that doctorsand lawyers competed for the best hous-es in Palo Alto. Now they’re all lawyers orventure capitalists; they’re not doctors.While there are a lot of good ideas in

flat tax reform, it wouldn’t be remotelypractical to do it with a single positivetax rate now. So I play around with sys-tems that have, say, two brackets. The“not-so-flat” tax. But of course thatdoesn’t have quite the simple appeal thatthe “flat tax” did. [Laughter]But there’s still a great idea in that book

which applies to any tax system, which is,it basically figures out how to implement

a value-added tax or other consumptiontax in a way that’s progressive.There were two economists on

President Bush’s Advisory Panel onFederal Tax Reform in 2005, JimPoterba and Ed Lazear, who reallyunderstood that. They pushed prettyhard; that was one of the designs thatwould make sense for how to do a con-sumption tax, even though it wouldn’tbe a flat tax.The origin of our initial flat tax effort

was Rabushka coming to me in 1980and saying, “I know what the peoplewant. The people want a flat tax, but Idon’t know quite what that is.” And Isaid, “I know what it is because I’ve beenthinking about it since I was a graduatestudent.” But, of course, for me, it was aconsumption tax—an efficient, simple,fair consumption tax. The flatness wasn’tso important but, of course, the flat taxname, which Rabushka contributed,was very important politically.

Region: Marketing is important.

Hall: Yes, but now the idea of tax flatnessis understandably not as popular.

DYNAMICS OF LABOR MARKETS

Region: You’ve also done a great deal ofresearch on labor markets. In 1982, youdocumented the “importance of long-term jobs” in the United States. I’m notsure that’s still the case.

Hall: It’s still the case. That paper’s beenreplicated quite a few times. It’s almost alaw of nature. The financial press is con-stantly telling us how much turnoverhas increased, how the old days of thelifetime job have disappeared. Butthere’s no particularly strong evidenceof that. There are some interestingchanges going on, but nothing that dra-matic.

Region: A 2005 paper of yours arguedthat job separation was also fairly stableand what was more important was look-ing at the hiring process and job find-ing.

Hall: That’s right.

Region: So you’ve been studying thatprocess carefully, looking at job searchdynamics, wage stickiness, wage bar-gaining, productivity, other factors.You’ve developed a model that explainslabor market fluctuations withoutassuming what you consider to be unre-alistically high labor supply elasticity.

Hall: I think “explain” might be a littlebit of an overstatement. I’m not surehowmany of my colleagues would agreewith the word “explain.” [Laughter] Ithink “accounting for” might be right.

Region: Fair enough. What factors haveyou found most successful in account-ing for job-finding rates? And what arethe key drivers of labor market volatility?

Hall: The important feature that con-trols the job-finding rate is the incen-tives to employers to create jobs. At anygiven time, if the incentives are not verystrong—it could happen for many dif-ferent reasons—then employers will dorelatively little to try to recruit workers.Job seekers will then have trouble find-ing jobs, will see themselves at the endof a long line of people waiting for thejob.Interestingly, the number of people

who find jobs each month is more orless a constant. Of course, this changes,but it’s a pretty good starting point forunderstanding labor market dynamicsthat the number of people who find jobseach month is the same in a strong mar-ket or a weak market.In a strong market, you have a rela-

tively small number of job seekers, soeach one finds it easy. In a good market,it takes the average person about amonth to find another job. In a weakmarket, there are twice as many peoplelooking, but each one of them is half aslikely to find a job each month; theproduct of the two—the number look-ing for a job and the fraction of themwho find a job—is the same.So, something like 4 million people

find jobs every month. Even with 10

The Region

33 JUNE 2010

Since 1981 when we first promoted that[flat tax] plan, there’s been a dramaticwidening of the income distribution inthe U.S. That means that the idea of thepoor paying the same tax rate just seemsless viable than it was when the incomedistribution was tighter.

percent unemployment, as recently,we’ve still seen the same thing. A verylarge number of people looking, verylow job-finding rate for each individual,but the product—the number of jobsfilled—is roughly a constant. It’s a veryimportant fact about the labor market.Think about a slack market from an

employer’s point of view. They see thereare all kinds of highly qualified peopleout there they can hire easily, so theydon’t need to do a lot of recruiting—people are pounding on the door.

Region: And these days they’re censustakers.

Hall: [Laughter] Right! So that’s the firstthing to think about: job creation incen-tives.If you ask, how did we get into a situ-

ation where job creation incentives havedeclined? It’s that there’s been a declinein the profitability of hiring a workerwithout a corresponding decline in the

wage. The incentive to create a job is thedifference between what a worker willcontribute to the business and what theworker has to be paid.That’s a very simple calculus. But that

seems to vary. In a recession, for variousreasons, the profit margin from hiring aworker declines, and that reduces job-creating efforts, all the things that keepthe labor market moving. And that, inturn, causes it to be difficult for the jobseeker to find a job.There’s a great debate going on as to

just what the factors are that reduce theadditional profit from hiring anotherworker.For a while, there was the thinking

that movements in productivity—pro-ductivity is one of the factors, so if pro-ductivity falls and the wage doesn’t fallwith it, then that reduces the profit mar-gin. But that idea has not worked in thelast three recessions because they wereperiods when productivity was rising,not falling. So the old productivity storyhas not worked for the last 30 years.But each of us has our own set of

ideas. To tie it to what we were talkingabout before, financial frictions havethe same effect. Increasing financialfrictions reduces the desirability ofadding workers. That’s especially true ifthere’s anything about the employmentrelationship that has an investmentcharacter. If a worker has to be trainedand becomes highly productive laborin time, then this question of what thecost of funds is becomes important. Arise in the cost of funds will result in adecline in employment, and that’ssomething a lot of people are looking atright now.There are many threads to this topic.

We’re debating actively which ones aremost important.

RECESSIONS AND RECESSIONDATING

Region: People are wondering when will,or did, the current recession end, but I’dlike to ask how you and the NBER[National Bureau of EconomicResearch] committee you lead decided

when it began. Many countries define arecession as two quarters in a row ofnegative GDP growth, and by that stan-dard I think the United States wouldhave entered its recession in, maybe, thethird quarter of 2008.

Hall: But that gets back to the wholequestion of, do you include the peak ofreal GDP?We always talk about the dateof the peak. That helps sort out this tim-ing. The peak occurred in the secondquarter of 2008. However, as you know,we declared the peak to be a little earlierthan that, December of 2007.

Region: Would you explain what stan-dards—I know it’s on the NBER Website; it’s very clear there—but could youelaborate on what standards you use todetermine turning points in businesscycles?

Hall: Actually, it’s not that clear, becausethese things are always up in the air.[Laughter] There’s a certain amount ofambiguity in what we put on the Website. We haven’t resolved some impor-tant questions about how this processshould work.

Region: Why really do we need a com-mittee, a dating committee, rather thanrelying on a rule of some sort, like twoquarters of negative GDP growth? I thinkyou’ve been on the committee since itbegan…

Hall: I’m the only chairman the commit-tee has ever had, for 32 years.

Region: I didn’t know you’ve chaired itthe entire time! Well then, you’re theright man to ask. Do you think it mightbe useful for the NBER, in addition todoing what it now does, to also issuesomething closer to a real-time indica-tor or signal of recessions—that couldbe revised for false positives or nega-tives, along the lines that Òscar Jordàhas recommended?

Hall: I think we feel that doing some-thing like that, and in any sense making

The Region

34JUNE 2010

Something like 4 million people findjobs every month. Even with 10 percentunemployment, as recently, we’ve stillseen the same thing. A very large numberof people looking, very low job-findingrate for each individual, but the product—the number of jobs filled—is roughly aconstant. It’s a very important fact aboutthe labor market.

it official, would somewhat cloud thingsbecause there would be enough type 1or type 2 errors [false positives or nega-tives]. We’re very happy to see that typeof research be done; we don’t claim anymonopoly on this point, and it’s beenvery instructive.Actually, long ago, in the 1980s, we

sponsored a project that informally,unofficially put out a recession proba-bility index that Jim Stock and MarkWatson prepared. It didn’t work verywell in the 1991 recession, so theystopped doing it after that.And it didn’t work for fairly typical

reasons. That was the first recession thatwasn’t accompanied by a decline in pro-ductivity, so it looked somewhat differ-ent. So their historical relationshipsweren’t as stable as they hoped.That’s one of the main reasons why

automatic rules haven’t worked. Peoplehave done research on the machineapproach for years. In fact, when I was agraduate student and took a computerscience course, my project was to writesoftware that would automate this. Soit’s not a new idea. But it’s never workedvery well.

Region: It would have missed the 1981recession if we’d used the two negativeGDP quarters rule.

Hall: You mean 1980.

Region: Right, 1980.

Hall: 1981 was no problem. The 1980recession was just one quarter. And peo-ple have said that the 1980 recession wasactually just sort of a prelude to the ’81recession. We say no, but it’s been said.

Region: It seems it’s more of an art thana science then.

Hall: It’s a classification problem that theworld seems to want an answer to, but ithas a shifting structure, and dealingwith the shifting structure is the issue.We try very hard to achieve historicalcontinuity.We don’t doubt for a second—and I

don’t think anyone else does either—that we know when there’s a recession.In all the data we look at, certainly inthe period when we’ve had reliabledata, which is since World War II,there’s never been an episode that’ssomewhere halfway between a reces-sion and a nonrecession. Every reces-sion has been clear. And they all seeunemployment shoot up and typicallysee GDP decline.We do face issues though. With the

most recent revisions of GDP, the 2001recession essentially doesn’t exist. It wasa flattening, but as emphasized on ourWeb site, there are issues of depth, dura-tion and dispersion, but there was nei-ther depth nor duration in what hap-

pened in ’01. By the alternative measureof total output, real gross nationalincome, the 2001 recession is quiteapparent.To me, it’s not an issue because that’s

just looking at GDP. If we look atemployment, as I did in a 2007Brookings paper on the “ModernRecession,”—by “modern recession” Imean one in which productivity rises…

Region: And monetary policy is undercontrol.

Hall: Monetary policy is stable, exactly.But here I think the key point is aboutproductivity. With rising productivity ina recession, you can see a relatively mildmovement in GDP, and there’s a longperiod of GDP growth at the same timethat employment is falling.When people talk about the jobless

recovery, it’s just another term for pro-ductivity growth. That’s complicated theprocess. The complication in the 2001recession is that productivity roseenough to offset employment declines,so we have a very pronounced, obviousrecession in employment and what’shardly a recession at all in GDP.I’m perfectly satisfied that’s a reces-

sion because I want to balance the two.To the extent you look just at GDP,though, it would be hard to call that arecession.That’s material today because, of

course, we’re seeing the same thing.GDP reached a very pronounced troughin the summer of 2009. It’s been prettyconsistently rising—with one little hic-cup recently—since then. So on thatstandard, we say the trough was in sum-mer of ’09 or maybe the fall of ’09.But employment is still declining. We

still have not seen a growth month.Everyone is presuming that we will inMarch—but that’ll be the first. You canplausibly make the trough of GDP be inJune of ’09, but the trough of employ-ment is probably going to be March of2010. That’s a long time.Not as bad as ’01, when the situation

was even worse. The trough in employ-ment didn’t occur until 2003.

The Region

35 JUNE 2010

Historical relationships weren’t as stableas ... hoped. That’s one of the main rea-sons why automatic rules haven’t worked.... It’s a classification problem that ... has ashifting structure, and dealing with theshifting structure is the issue. We try veryhard to achieve historical continuity.

The complication in the 2001 recessionis that productivity rose enough to offsetemployment declines, so we have a verypronounced, obvious recession in employ-ment and what’s hardly a recession at allin GDP. ... That’s material today because,of course, we’re seeing the same thing.

STOCK MARKET VALUATION

Region: Let me ask about the stock mar-ket. Roughly a decade ago, you did a lotof work on eCapital, eMarkets and stockmarket valuation. Your 2001 Richard Elylecture was an example of that. And yousuggested that investors did seem to befairly estimating the market’s value ifintangible capital was taken intoaccount. Is that accurate?

Hall: Well, I talked about some individ-ual cases where I thought you could tellthe story. There was a discussion of eBayin the Ely lecture. On the other hand, ifyou look at the results in my AER[American Economic Review] paper, itobserves that intangible capital by thatmeasure was deeply negative in the mid-’70s to about 1980. Now, positiveeCapital makes a lot of sense, but nega-tive eCapital is a little hard to swallow.So I’d be careful.There was something weighing down

the stock market from basically 1974 to

1990. eCapital turned positive in 1990.So during that period, there was someundervaluation. It was very clear thestock market later decided it was anundervaluation because if you made astock market investment in 1980 andheld it to 1999, you had a very largeexcess return in the 20-year period. So Ithink there are still some mysteries.In spite of the fact that the valuation

that we see in the market right nowseems to be in a reasonable range, thereturns since 1999 have been way belowany benchmarks, which suggests thatthere was some overvaluation then.

INTELLECTUAL PROPERTY

Region: You’ve thought and written agreat deal, in both technical and laypublications, about the economics ofcomputers and software, as well as ven-ture capital and entrepreneurs. Thatseems natural given that you were bornin Palo Alto and have worked here for along time.

Hall: Flora Hewlett, married to theHewlett of Hewlett-Packard, was myfather’s secretary when he was aStanford professor. If only he’d boughtone share!

Region: You’ve also devoted some timeto studying antitrust economics, andlooking at potential for monopoly pric-ing in upstream supplier markets.What is your view of the argument

that intellectual property, copyright lawsand patents inhibit rather than encour-age innovation?

Hall: First of all, I think that that’s onlybeen directed at patents. I don’t thinkthere’s any feature of copyright law. Itprotects the expression. There’s an infi-nite space of melodies that composerscan compose and once they do, it doesn’tinhibit other composers from compos-ing other songs because there’s this infi-nite space. Every expression is complete-ly unique, so when it comes to expres-sion, I don’t think there’s any real issue. I

The Region

36JUNE 2010

More About Robert Hall

Current Positions

Robert and Carole McNeil Joint Senior Fellow, Hoover Institution,and Professor, Department of Economics, Stanford University; joinedfaculty in 1978

Previous Positions

Massachusetts Institute of Technology, Professor, 1970–78

University of California, Berkeley, Assistant Professor, 1967–70

Professional Affiliations

President, American Economic Association; President-elect, 2009;Vice President, 2005; Ely Lecturer, 2001

Director, Research Program on Economic Fluctuations and Growth,National Bureau of Economic Research, since 1977; Chairman,Committee on Business Cycle Dating

Member, National Academy of Sciences, since 2004Member, Advisory Committee, Congressional Budget Office, since 1993

Member, Oversight Panel for Economics, National Science Foundation,1989; Advisory Panel for Economics, 1970–72

Honors and Awards

Fellow, American Academy of Arts and Sciences, Econometric Societyand Society of Labor Economists

Hall of Fame, Money magazine, with co-author Alvin Rabushkafor their book The Flat Tax

Publications

Author of Forward-Looking Decision Making: Dynamic ProgrammingModels Applied to Health, Risk, Employment, and Financial Stability(Princeton University Press, 2010), Digital Dealing: How eMarkets AreTransforming the Economy (Norton, 2002), The Flat Tax (with AlvinRabushka, Hoover Institution Press, 2d ed., 1995) and The RationalConsumer: Theory and Evidence (MIT Press, 1990), among other books.Widely published in academic journals, with research focused onemployment, technology, competition and economic policy

Education

Massachusetts Institute of Technology, Ph.D. in economics, 1967

University of California, Berkeley, B.A. in economics, 1964

think almost everyone believes in apretty powerful IP rights regime forexpression.When it comes to the things that

patents protect, then the patent regimehas to do the things that the patentregime claims to do. The patent has tobe original; it has to be an innovation.And there the standard of obviousnesscomes in.If what’s happening is that people are

somehow able to figure out what theobvious next logical step is and some-how get a patent on that and then collectroyalties from that patent even though itdoesn’t really make any contribution,then there’s something wrong with thepatent regime. But I don’t think there’sany very good evidence that that’s actu-ally what’s happened.People make fun of a lot of the

patents that the patent office issues, butthey don’t matter. There’s only a muchsmaller set of patents that have everattempted to be enforced and have

caused any problems. On the otherhand, the patent system has generatedsome very substantial rewards to sometrue innovations.You know, it’s all in the details. I don’t

accept a broad condemnation of thepatent system. I don’t join any of thesepeople who say there shouldn’t be anybusiness process patents or thereshouldn’t be software patents. Somegood ideas are implemented in software.What is a good idea, and what every-

one stands by, I think, is the notion thatpatents shouldn’t last forever. The ideaof a finite patent life, which is currentlyaround 20 years, does seem to be animportant part of the design.The result of that is that the great

majority of innovations ultimately ben-efit workers in the form of higher wagesrather than any permanent stream ofmonopoly profits going to owners. Ifthat weren’t true, you’d see a hugeamount of innovation value capitalizedin the stock market, but you don’t, andthat’s proof. Consistent productivitygrowth and corresponding real wagegrowth is demonstration then that thebenefits ultimately of innovation aregoing to workers. So it’s a great thing.

THE STATE OF ECONOMICS

Region: The past few years seem to havebrought about a crisis of confidence inthe economics profession, with criticssuggesting that macroeconomics hasfailed in some fundamental way. It’s atopic addressed by [Minneapolis FedPresident] Narayana Kocherlakota inour Annual Report this year. Do youagree that the macro profession failedthe nation during the financial crisis?

Hall: I don’t. There are two parts to theissue. First, did macroeconomists fail tounderstand that a highly levered finan-cial system based in large part on real-estate debt was vulnerable to a declinein real-estate prices? No way. Many of uspointed out the danger of thinly capital-ized banks. We had enthusiasticallybacked the idea of prompt correctiveaction in bank regulation, so that banks

would be recapitalized well before theybecame dangerously close to collapse.We watched in frustration as the regula-tors failed to take that action, eventhough they had promised they would.Second, did macroeconomists fail to

understand that financial collapsewould result in deep recession? Not atall. A complete analysis of that exactissue appears in an extremely well-known and respected chapter in theHandbook of Macroeconomics in 1999,written by Ben Bernanke, Mark Gertlerand Simon Gilchrist. Depletion of thecapital of financial institutions raisesfinancial frictions to levels that distinct-ly impede economic activity. In particu-lar, credit-dependent spending on plant,equipment, inventories, housing andconsumer durables collapses. Thatchapter is an excellent guide to thedepth of the current recession.

Region: Thank you for a great conversa-tion.

—Douglas ClementMarch 16, 2010

The Region

37 JUNE 2010

Patents shouldn’t last forever. The idea ofa finite patent life, which is currentlyaround 20 years, does seem to be animportant part of the design.

The result of that is that the greatmajority of innovations ultimately benefitworkers in the form of higher wages ratherthan any permanent stream of monopolyprofits going to owners.