Why do economists disagree?

Non-economists are often baffled by the disagreements among professional economists on the issues of the day--from international trade to the minimum wage, from economic development to health policy.

I think the best way to understand the source of these disagreements is to recognize that there are two genres of economists. I call them "first-best economists" and "second-best economists." Here is my guide to them.

You can tell what kind of an economist someone is by the nature of the response s/he offers when confronted with a policy issue. The gut instinct of the members of the first group is to apply a simple supply-demand framework to the question at hand. In this world, every tax has an economic deadweight loss, every restriction on individual behavior reduces the size of the economic pie, distribution and efficiency can be neatly separated, market failures are presumed non-existent unless proved otherwise (and to be addressed only by the appropriate Pigovian tax or subsidy), people are rational and forward-looking to the first order of approximation, demand curves always slope down (and supply curves up), and general-equilibrium interactions do not overturn partial-equilibrium logic.  The First Fundamental Theorem of Welfare Economics is proof that unfettered markets work best.  No matter how technical, complex, and full of surprises these economists' own research might be, their take on the issues of the day are driven by a straightforward, almost knee-jerk logic.

Those in the second group are inclined to see all kinds of complications, which make the textbook answers inappropriate. In their world, the economy is full of market imperfections (going well beyond environmental spillovers), distribution and efficiency cannot be neatly separated, people do not always behave rationally and they over-discount the future, some otherwise undesirable policy interventions can generate positive outcomes, and general-equilibrium complications render partial-equilibrium reasoning suspect. The First Fundamental Theorem of Welfare Economics is proof, in view of its long list of prerequisites, that market outcome can be improved by well-designed interventions. Since they have given up on the textbook model, members of this group have an almost-infinite variety of "models" to choose from as they think of public-policy issues.

The first group's instinct is always to apply the first-best reasoning to the case, ignoring market imperfections in related markets, while the second group almost always presumes some market imperfections in the system. I am over-simplifying a bit, but not a whole lot.

Among commentators in the blogosphere, I think Gary Becker, Tyler Cowen, Greg Mankiw, and Brad De Long (more often than not) are first-best economists. In is commentary on globalization, Jagdish Bhagwati is an unadulterated first-best economist, even though his best scholarly work is solidly cast in the second-best mold. Meanwhile, the undisputed king of second-best economists is Joe Stiglitz. He is joined by George Akerlof, Bob Shiller, Alan Blinder (recently) and Paul Krugman (especially when he writes on deregulation and health policy, and increasingly, but not always, on trade). I am definitely in the second-best camp as well.

When first-best economists are taken to task for ignoring real world complications--i.e., second-best interactions--they provide a range of answers. One is to downplay the significance of these issues by arguing that they are not convinced of the presence of the market imperfections in question. Sure enough, empirical evidence is hardly ever strong enough to move prevailing priors.

A second argument is that the presence of additional market imperfections does not change the first-best logic; it simply calls for each market imperfection to be treated with its own first-best solution. This allows each expert in a field to propose first-best solutions in that field, leaving complications elsewhere to be dealt with by others. Larry Summers had a nice point to make about this approach in his comments on a paper on banking reform in China (Brookings Papers on Economic Activity, 2006:2):

Like experts in many fields who give policy advice, the authors show a preference for first-best, textbook approaches to the problems in their field, while leaving other messy objectives acknowledged but assigned to others. In this way, they are much like those public finance economists who oppose tax expenditures on principle, because they prefer direct expenditure programs, but do not really analyze the various difficulties with such programs; or like trade economists who know that the losers from trade surges need to be protected but regard this as not a problem for trade policy.

(Come to think of it, is Larry Summers a first-best economist or a second-best economist?)

A third argument is that the government could never get complicated interventions right, so we are better off sticking with simple solutions. I have discussed this type of argument in an earlier post.

So at the end of the day, these disagreements are often grounded not in economics per se, but in strongly held prior views about the world in which we live in. Which is why non-economists are right to get exasperated with us.

UPDATE: Bill C. reminds us that Keynes had some very apt things to say in his General Theory on this very same distinction. His fight at the time was with the Classicals. Here is Keynes, via Bill C.:

The completeness of the Ricardian victory is something of a curiosity and a mystery. It must have been due to a complex of suitabilities in the doctrine to the environment into which it was projected. That it reached conclusions quite different from what the ordinary uninstructed person would expect, added, I suppose, to its intellectual prestige. That its teaching, translated into practice, was austere and often unpalatable, lent it virtue. That it was adapted to carry a vast and consistent logical superstructure, gave it beauty. That it could explain much social injustice and apparent cruelty as an inevitable incident in the scheme of progress, and the attempt to change such things as likely on the whole to do more harm than good, commended it to authority. That it afforded a measure of justification to the free activities of the individual capitalist, attracted to it the support of the dominant social force behind authority.

As usual, Keynes puts it best.

UPDATE2: The link in the following paragraph was incorrect; I have now corrected it:

A third argument is that the government could never get complicated interventions right, so we are better off sticking with simple solutions. I have discussed this type of argument in an earlier post.

It is highly relevant to Tyler Cowen's response.

 

Comments

[gabriel]

I think your analysis is spot on. I am definitely in the second best camp after being in the first best camp after receiving my Phd. However, having studied development economics and now having spent 7 years in professional consulting, the markets are full of imperfections that would cast doubt on first best economics. However, in a pinch, when pressed for time, it is amazing how fast we all resort to that first best world!

Yet, I am certainly in the second best camp.

[Charles]

I think as a representation of two poles on a continuum this is accurate. However, given the complexity of the problems economists often analyze it is clear that some sort of model/heuristic is required.

These economists generally choose a model or heuristic which complies with their view of the world. Using decision theory, which camp is most likely to be making the optimal recommendation? Which one has the greatest cost of being wrong in their particular area? This could be viewed as a Bayesian estimation problem viewed under a decision theory paradigm.

Neither camp will be right all the time. But thinking about the policy problem like this is a sensible approach.

[Krishna]

Appears similar to the rules/principles vs. discretion distinction, with the first-besters in the former camp. Relatedly, one's view may also depend on the time frame over which you are analyzing policy and the confidence you have in economic methodology to generate specific, falsifiable predictions as to the impact of particular interventions.

[unarmed]

It's instructing to see that most first-best economists mentioned here (Cowen, Mankiw, etc.) could be, in a nutshell, classified as right-wingers, while almost all of the second-best economists (Stiglitz, Krugman, Rodrik?) would be very comfortable if placed under the left side of the political umbrella.

Now the timeless question is this: is political standing a cause, a consequence, or both of these guys' economic thinking?

[gabriel]

Hmm..not sure about whether they are all "right wingers". I don't think one would classify Delong as a "right winger" using the more popular definition of "right wing".

One thing to note, is that both Akerlof and Stiglitz may be seen as "second best" economists because they each spent time early in their careers in developing countries (i.e., India). However, one must keep in mind that both also spent a lot of time looking at Growth models in the 60s. See the book by David Warsh, "Knowledge and the Weath a Nations".

[jb]

I have the impression that Mankiw is in the first-best camp when talking to undergraduates and non-economists (thus, in his blog). However, some parts of his research are on market rigidities (menu costs, sticky information and others), a feature of the second-best camp. Using this definition, most New Keynesians (MIT-Keynesians) actually fall in the second category, right?

Those "first-best" people always remind me of Doctor Pangloss. Someone ought to rewrite Voltaire's "Candide" to the economist setting.

[Robin Hanson]

I think both sides agree there are lots of big imperfections. The difference seems to be more in judgments about our net ability to identify which particular imperfections exist how much in which particular situations, to craft appropriate remedies, and to get political processes to adopt those remedies. Those who think our abilities end up being poor will think attempts at fine tuning are more likely to make things worse; those who think our abilities are good will lament lost opportunities to make improvements. So what evidence could we collect that would help us to differentiate between these, by better estimating our net abilities?

[Dan Olner]

This question about the interaction between politics and economic theory is interesting / complicated. Most economists would, I think, agree that there is such a thing as an objective answer to economic questions. Its only non-economist lefties who argue that economics is nothing more than a pseudo-scientific figleaf to cover the ideological nature of its roots.

But then, imagine this scenario: in Fictionland, there are a hundred different takes on any economic question. The people who run Fictionland are a mix of the rich and powerful: its a two-party system, but each consists of a not-so-different set of these rich / powerful social classes. What kind of economist is likely to be invited into the room when the economic policy decisions are being made? I'd argue its going to be the ones whose theory supports the continuation of the current group of people in power. After all, history is not littered with examples of people with power willingly giving it away.

So I'm saying the empirical validity of a theory is not the main factor in its becoming accepted into policy circles. Its an evolutionary interaction between power and theory: the theory that survives in the landscape of power wins out. Its empirical strength is a factor in its success, but not the only one, and perhaps not even the most important.

One US democratisation theorist says: 'a democratic polity informed by a genuine commitment to markets is in the long run the best political response to the problems posed by the rational expectations of economic actors... When the opposition supports the basics of a market economy, actors can rationally expect that the end of a particular administration will not spell the reversal of all their economic expectations.'

This makes me wonder: what's the best way of representing the diversity of economic opinion in a polity? The above kind of democratisation theory (and the actions of the World Bank / IMF in developing countries) pushes the notion that economics should be kept separate from democracy. I'd argue the reverse: its inevitable that different economic theories benefit different interest groups, and the only way we've come up with to manage this kind of conflict is democracy. Exactly what kind of democracy is most appropriate for deciding economic questions, God only knows. Countries that are currently liable to find themselves having to choose between the World Bank or no aid at all should have more choices, though - and that choice should be treated as a serious democratic decision for that country to make, not a technocratic choice of the best policy, to be made by the country's finance ministers and a bunch of faceless World Bank bods who've just flown in the night before.

There are so many problems with this approach, though, that I don't know where to start. But I think this one one reason why 'first-best' economists cling to orthodoxy. Once you've accepted there's diversity of economic opinion, its much harder to argue that economic decisions should be made in isolation from the democratic process. This is the exact opposite of the rationale for e.g. the WTO - which has been described by two of its proponents as a mast to which governments can tie themselves, to "escape the siren-like calls of various pressure groups." Bah.

[david]

It's the idea of "imperfection" that makes the two camps so difficult to distinguish. Oh, if only!

That, and power.

[david]

It's the idea of "imperfection" that makes the two camps so difficult to distinguish. Oh, if only!

That, and power.

[david]

It's the idea of "imperfection" that makes the two camps so difficult to distinguish. Oh, if only!

That, and power.

[Bill C]

Very insightful. Reminds me of what Keynes said in the intro to the General Theory: "...the postulates of the classical theory are applicable to a special case only and not to the general case, the situation which it assumes being a limiting point of the possible positions of equilibrium. Moreover, the characteristics of the special case assumed by the classical theory happen not to be those of the economic society in which we actually live, with the result that its teaching is misleading and disastrous if we attempt to apply it to the facts of experience." Sixty-one years later, we're still fighting the same battles!

[Lee A. Arnold]

Thank you for this post because I have always wondered how the first-besters' refutation of the theory of the second best is executed. I expected more -- it's a bit of an intellectual letdown. The refutation of a theoretical result by an imagined series of contingent corrections does not seem scientifically respectable, while the actual number of necessary corrections in a real economy would make it practically unachievable. Are the first-besters dancing with faith? Indeed it makes the first-besters into crusaders for perfection everywhere -- something that political conservatives (at least in the U.S.) usually find to be less than brilliant, although then the discourse is forgotten and subtly shifted to the crusade for freedom everywhere. But it is not entirely clear to me that free markets, love them as we do, always increase freedom. Consumer choice takes time to perform, and in difficult questions, such as choices among basic healthcare policies, it may take perhaps more knowledge than many people can have. Time consumed is other time denied, and so there are certain markets which may not increase freedom. Or may cost more, because expert information has been artificially made necessary, and then it must be purchased. These objections are too readily glossed over; or the argument is again shifted mid-sentence back to efficiency -- for which the first-besters must fall again upon their patchwork.

[Joe]

Excellent! (Interesting that Mankiw is currently dissecting economists in his blog!. But with a twist.)

But I really wonder if Jagdish Bhagwati is truly an "unadulterated" first-best economist.

On the question of trade, yes, he is. But on the issue of Globalization of finance he is not - in fact, I haven't been able to see any difference in the views of him versus those of "the undisputed king" of second-best economists: Joe Stiglitz; Like Stiglitz, Bhagwati has long been an advocate of capital controls in developing countries.

[R Mutt]

Another possibility. I think when it comes to politics, the first order economists think that politicans are biased in favour of bad economics, and that the basics need to be stressed to stop giving the politicians justifications.

For instance, politicians don't care about hidden economic costs: if it's hidden it doesn't cost votes. Politicians care more about concentrated effects than distributed effects: distributed effects tend to be too small to influence the average voter. Politicians don't really care about the long term: boosting growth by a percent will have a huge effect over decades, but doesn't help them win the next election.

I think the first-order economists see politicians as constantly looking for justifications for unnecessary market interventions, and they think that politicians will use the second-order effects as an excuse.

The question is: which group is more politically savvy?

[Econ_HH]

Dani Rodrik made a good point in making a distinction along the lines of an economist's position towards the first fundamental welfare theorem.
However, there are two other dimensions which matter.

1. What is the economists view on human behaviour? Put it differently: What is the economist's a priori view on the embeddedness of economics into the society as a whole -- including norms, institutions, social networks. To give an example: Brian Caplan in The Myth of the Rational Voter analyzes the non-rationality of voters in a nice way but ending in the proposition to make the voter more rational. What are we as economists? God? Should'nt we respect other human beings and try to explain instead of changing the way people see the world?
2. Mankiw's view on the economist as a Scientist or Engineer. That is another important dimension which distinguishes the public policy crowd from the ivory tower.

By the way, what kind of economist was JM Keynes?

[aje]

Two types of economist, those who see the first fundamental as the best way to classify economists, and those that don't.

I agree with Robin, the issue isn't estimates over the size of imperfections, it's assumptions about the knowledge requirements to "fix" them.

[Christopher]

I don't think you quite do the first-best crowd properly. Instead of saying that the 'market' will work perfectly, a more accurate depiction would be:

first-best: Believe that people act within the context of a market more intelligently than they do within the government.

It's not that first-best economists don't think the government could intervene (or that it hasn't intervened in the past) in a way that is productive, but rather that the structure of government lends itself to bad decision-making, or at least worse decision-making than a functioning market.

There is dispute about how big of a role the gov't should take in preserving market functionslity, and I think 'second-best' economists have a lot of insight to offer here. Still, let us not forget that the principle division between the first- and second-best economists lies in faith in the government.

[op]

if u start
by figuring out who seems to benefit from a policy
then trying to crack open
the rationalization
as rationalization
most times
"cause and result "
becomes obvious
whether the science-izing
is performed
by a useful idiot pangloss
a ass-burger gradgrind
or ...greg mephisto mankiw
and his uncle marty

correcting
for the mistakes
in calculating
group interest
the eternal veil
of ignorance
and stupidity most assuredly produce

the motive is revealed
in the expected result ahead
not the reasoned
science behind

[op]

chris:

"the structure of government lends itself to bad decision-making, or at least worse decision-making than a functioning market"

this evades
the whole question
of general welfare

benthamite
"look to one's self interest "
has well defined limits
even for benthamites

market failure
is a basis
for collective choice

gubimnts are self correcting too

in the long run
gubmint is not
an unnecessary evil

ps
general welfare improving
changes
by state intervention
may easily
move market activity
away from inferior
spontaneous outcomes
---at least at the margin---

and
such interventions
may often require very little
"information"
only
the mobilization
of a sufficiently
"larger "interest
to overcoming
the pack of winners
now being
rewarded by spontaneity

[CityEconomist]

A letter to the NY Times a couple of weeks ago put it well, something like this (from memory):
"There are two kinds of economists:
1. Those who believes that markets always work, except for when they don't.
2. Those who believe that markets never work, except when they do."

[Peter]

This has been an interesting exchange. One thought that comes to me from work in benefit-cost analysis is this: the fundamental division is between economists who see the political system primarily as a source for error, and those who have enough faith in democratic politics to see themselves as contributors to it.

The political skeptics tend to emphasize the corruption and irrationality found in all known systems, and they point to the theoretical problems with decision by majority rule. The role of BCA, in their view, is to remove as many decisions as possible from the political arena, and have them determined by whether B>C or C>B. This entails a strict adherence to BCA formalisms, including the monetary valuation of all outcomes of a policy choice.

Believers in democracy see BCA as an input into a larger process that may confer a greater degree of wisdom and social acceptance than economic analysis alone can provide. This means that the bottom line of the study is not the bottom line of the policy process, so strictures can be adhered to selectively. In particular, one may choose not to monetize a particular result (for instance health) or to point to cultural or political factors that economics may sidestep altogether.

The practical consequences of these two positions in the way analysis is conducted are profound. They depend, however, on priors that are nearly impervious to rational argument. You can show me (a believer in democracy) case after case of democratic political failure, and I will still cling to the hope that the next time around it will be different. Anti-political economists, who I think constitute the majority, will be unimpressed by any example of democratic success. How can we be sure, they will ask, that this will happen next time?

The link to the first-best/second-best dichotomy can be found in welfare economics. If you believe, in general terms, that orthodox welfare econ really speaks to human well-being, you will tend to place more stock in the rest of the argumentation that links efficient market (and pseudo-market) mechanisms to welfare maximization. But, on my reading, welfare economics has walked off an empirical cliff some time ago. The evidence on human behavior and well-being is profoundly incompatible with the basic premises of the received theory. It would be interesting to investigate the sociology of knowledge aspect of this situation, the delayed effect this development is having on the profession, but I think that on logical grounds, once the awareness of what has happened to welfare econ has seeped in, it will be much more difficult to justify the anti-political perspective.

[hari]

Dani is perhaps raising a hidden problem; namely, are economists becoming a serious problem in govt decision-making on bread and butter issues? It'd seem, for us +70 old timers, that both politics and economics has moved along, and globalization (gatt/wto) has made the average citizen reticent about those who control levers of power in govt.
They don't seem to understand fundamentals of economic decision-making which becomes ever more complex and global in terms of its impact.
For those of us who have professionally spend more than three decades trying to prune down tarrifs and trade under gatt rounds, the outcome is not very flattering-at all!
The imperfection of economic science is further eroded by its intrinsic inability to provide simple answers.
My take on "first" and "second" group is simply a function of US politics. It may not be the same outside US. Which leads me to argue that perhaps we need more "political economists" to come into grips with evolving problems of globalization.

[robertdfeinman]

Perhaps other divisions might make more sense:
optimists/pessimists
humans are selfish/humans are altruistic
humans need strong leaders/humans can govern themselves

I suspect that the psychological outlook of many people makes them lean towards certain economic and social viewpoints. They then tend to find information which reinforces their viewpoint and slight that which doesn't.

Since much social and economic data is ambiguous this gives lots of scope for mischief.

[Bruce Wilder]

I am not as interested in the dichotomy, as in the relationship to research experience and results, which you posit.

Paul Samuelson's Foundations of Economic Analysis codified economic theory in a way that makes the reflexive return to the first month of Econ 101 (as Atrios described it) a natural first response. It was an incredibly important achievement for economics, and was only reinforced by Arrow-Debreu, but it privileged theory over all else in economics.

What economics desperately needs is some kind of framework, to catalog empirical experience and results. Instead, even your second-best economists are likely to respond with a meandering narrative thru a catalog of theoretical models, rather than a deep knowledge of institutions and circumstances.

In trade, the Heckscher-Ohlin model persists as a frame for thought, despite abundant empirical evidence of its limitations and deficiencies. It is the modern triumph of Ricardo over reality.

The dominance of case study in biology might be a good model for reform. Biology has a strong theory, in Darwinian evolution, but Darwin came along long after Linnaean classification and empirical results dominate, in part because an elaborate catalog organizes empirical results as examples of variety and process.