Two ways of thinking about economics

http://econlog.econlib.org/archives/2015/03/ways_of_thinkin.html

by shawndumas • 11 years ago

58 54 11 years ago

econlog.econlib.org

bandrami3• 11 years ago

Isn't Krugman just being scientific here? ie, "We had this model for a long time, but the empirical data don't back it up anymore so it's pretty silly to stick with it, and we should find a better model."

carapat_virulat• 11 years ago

Yes, I really enjoyed the article but I am not sure what was the conclusion. Should we respect the laws of economic theory even if they have no empirical influence in the real world? Or was Krugman just acting on faith and he was finally convinced by hard data, but it took some overcoming of his bias?

cromwellian2• 11 years ago

If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence. I used to be a hardcore libertarian, but all of the horrors I used to believe over marginal tax rates or regulation didn't materialize, moreover the blatant market failures and externalities both positive and negative are too hard to ignore.

When you are a hard core Libertarian or Objectivist even a modest tax increase or bump in CAFE standards seems like creeping communism to be resisted at all costs.

When I started looking at actual data across many countries I really couldn't see a correlation between my beliefs and the outcomes I desired. Many countries economic success or failures seem related to other factors outside the tax or regulatory regime.

E.g. Silicon Valley is a bastion of innovation but also high is costs, taxes, and regulations, with a politics structure that is essentially a basket case. That's because the nexus of people and network effects and culture are more important than the tax regime unless said regime is crazy bad.

Ergo, Krugman is somewhat right -- it's about people and structure (his work on international trade and industrial nexii).

Too much obsession over taxes and regulation "amount" rather than on what's contextual right is what's wrong with economic debate today.

logicchains4• 11 years ago

If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence.

Many classical and neoclassical economic propositions are of a fundamentally epistemological nature, not strictly positivist. Such as the reasoning that, if without interference person A does action X at time T, X may be taken as the action they most preferred to do at time T, as if it wasn't then why would they have done it? In which case any interference with their taking of action X necessarily produces an inferior outcome in terms of their preferences at time T, as they're forced to take an action other than their most-preferred action. In this sense a person's preference for an action is defined by their taking of that action over all other possible ones when not interfered with, and as a definition its validity is a matter of epistemology, not something subject to empirical analysis. The Austrian school uses this ordinal form of utility as its basis, building upon the idea that interfering with someone's free action is the one thing that can clearly be shown to be undesirable as above.

The Chicago school, on the other hand, also incorporates a quantitative form of utility, to allow for interpersonal utility comparisons. This requires some epistemological assumptions, for instance that preferences don't change over time or their rate of change follows a particular mathematical function, or that if someone buys 3 apples for $30 they value each at $10, rather than e.g. the first at $13, the second at $2, and the third at $15. Austrians dispute the validity of any such assumptions, and hence reach different conclusions.

These epistemological foundations are matters of logic/philosophy, and hence no more amenable to empirical analysis than a proposition like 'abortion is immoral'.

Modern economics tends to ignore epistemological issues, and instead focuses on finding the best way to maximise particular empirical metrics, without going into detailed study of the epistemological and philosophical validity of the metrics in question. So it's important to distinguish between economics as a tool for predicting and assessing the effects of particular phenomena on particular metrics (econometrics) and economics as an epistemological approach.

notahacker• 11 years ago

Some of it's that and some of it's the fact that he's addressing different arguments, and from somewhat shifted political perspectives

Both Krugman's quoted passages acknowledge that conventional economic theory would tend to suggest that minimum wages reduce the size of the work place, but both labour economists' theories of labour supply and some empirical work suggests that this isn't universally the case, especially where increases are small. Neither say very much on the subject.

But the 1998 one is attacking a book which (according to Krugman) suggests that compelling employers to pay more - 57% more in the example - will reduce welfare costs and have no adverse effect on employment

The recent article attacks a more general view that wage reduction is essential to get the nation's economy out of a rut, and is particularly approving of quasi market-based mechanisms to raise minimum wages (trade unions and Walmart's justification in terms of efficiency wages)

It's the political line of argument rather than the use of economic theory that has changed (and given that Krugman doesn't exactly pull his punches, it's not surprising that one article reads almost like an attack on the other). He's not having to turn his back on economic theory to argue that the US would be better off if minimum wages were a little higher, and it wasn't slavish devotion to economic theory that caused him to contest what he described as a "moral" argument for wage regulation over redistribution from Pollin and Lice.

bsbechtel4• 11 years ago

If you talk to any economist that does research for a living, they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable. Chief among the reasons for this is that it is impossible to collect data on all the different factors that go into a human's decision making process, and also that there is an observer effect (similar to that in physics) that influences data collection in many types of economic studies, especially forward facing ones.

bandrami3• 11 years ago

they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable

Then economics absolutely needs to stop being called a "science".

xixi772• 11 years ago

the point of the article is that this is what the author thought was the case, but not what actually appears to be the case -- e.g. it seems that Krugman's interpretation of the same Card & Krueger evidence has changed, as opposed to him changing his view based on new empirical evidence available since then (which is quite messy and inconclusive, tbh).

bandrami3• 11 years ago

In physics, when a new study is released physicists don't change their opinions overnight. There's a big difference in how a scientist treats a study that was recently released and one that has survived a decade or so.

lumberjack• 11 years ago

Will there ever be an article on econlib.org that says "free markets are not very competitive in such and such cases and maybe regulation is needed in such and such markets"?

I don't want to be inflammatory but pretty much all economic blogs are intellectually dishonest junk where economists, Nobel Prize winners and all, write simplistic pop-economics supporting the political leanings of whoever funds their research.

I'm somewhat bothered that these blogs, make it to the top of HN. Surely smart people should be bothered by the intellectual dishonesty, even if it might reaffirm their beliefs.

kfk• 11 years ago

What an arrogant way to see things. All bloggers economists write "intellectually dishonest junk" now. And talking about simplistic, isn't your comment too simplistic? It puts together everything and everybody in 1 box, you can't go more simplistic than that.

tsax• 11 years ago

Assuming automatically that the state stands outside the nature of man as a perfect institution with the ABILITY to 'correct' market outcomes. There's whole fields of literature examining this and most regulation falls far short.

dimitar2• 11 years ago

Is this post really about two ways of thinking or is it about the complex feelings Sumner has for Krugman?

Sumner doesn't seem consider how the minimum wage interacts with Earned Income Tax Credit and that the minimum wage is essentially unchanged since that era - what was high then could be low now. "How much?" matters.

There is also new theory since then, much more empirical evidence than the few flawed studies, and of course it is OK to change your mind.

ArkyBeagle2• 11 years ago

I am pretty sure that Sumner has stated a preference for the EITC over a minimum wage over time. As I understand it, that's from a purely control-theory perspective - a lower bound on wages may or may not reduce actual employment so much as it causes a loss of resolution in price signalling. I personally find the whole question gloriously undecidable. Minimum wage exists as a sort of "dead band" in labor markets.

I suspect the debate ( such as it is ) between the Monetarists and... whatever Krugman is now is the most productive thing in economics. It's about what and how we measure.

The last sentence is the money shot. "Krugman is...the economist for people who hate economics". Just my guess here, but I think that has much more to do with whether or not the Efficient Markets Hypothesis holds over what domains. In a post-Lehman world...

AnimalMuppet2• 11 years ago

I believe that Krugman's position is "market monetarism", but I could easily be mistaken...

dimitar2• 11 years ago

Scott Sumner and other monetarists generally like Krugman for his view on monetary policy. They even co-star in comics by fellow blogger David Beckworth:

http://3.bp.blogspot.com/-4xn6oZYpEMI/VYLQtGGFhiI/AAAAAAAASZ...

http://4.bp.blogspot.com/-zvNRPOtoIjM/VYLQ1-n-CAI/AAAAAAAASZ...

bsbechtel4• 11 years ago

Given he's an economist, I wish the author would consider the microeconomics surrounding Krugman's shift in viewpoints. Krugman has gained considerable influence and wealth by promoting left-wing economic ideas in his NYT column. While Krugman may have genuinely changed his mind, writing about ideas that speak to the hearts of millions of readers each week and make you lots of money will influence your writing, even if you're not consciously aware of it.

Nimitz14• 11 years ago

http://www.newyorker.com/magazine/2010/03/01/the-deflationis...

auntienomen2• 11 years ago

Krugman's gained influence and popularity by writing for the NYTimes, yes, but I doubt the gig pays him more than Princeton did.

saosebastiao• 11 years ago

I would be willing to bet a newspaper pays dearly for the prestige of having a Nobel Laureate on its staff.

kephra• 11 years ago

I wonder, why he is ignoring the iron law of wages: Under free market conditions, wages tend to the minimum necessary to sustain the life of the worker.

When free market is interrupted by state, union power or lack of workers with a special skill, the nominal wages rise. But the situation for the workers will stay same, because of the Iron Law of Wages constrain. There will still be some days left end of the month, when he runs out of money. Because when wages rise at a local market, the housing cost will rise next at this market.

This can be seen best in Switzerland where wages, taxes, and housing costs differ in every canton. Similar if you compare housing costs and wages in upper state New York, and in NYC. Housing costs and taxes will rise, if wages rise. Pushing those with the lower wages out of the market of better housing.

So if a state wants that minimum wages or unconditional income grants have a positive effect on the people, then the same state must invest in social housing first, to lower the housing costs, before raising wages.

federicobond• 11 years ago

Iron law of wages? I thought that debate was over a hundred years ago, together with all the Malthusian doomsday predictions. I guess I was wrong.

furyg32• 11 years ago

Watching economists claim to be amoral is amusing when they so transparently have moral beliefs they refuse to acknowledge.

Valuing unregulated markets over other priorities is a moral philosophy, one that is meaningless since there is no such thing as a completely free market. Claiming you know what the result of a huge national policy like raising the minimum wage based on the extremely limited data of the field is disingenuous, at best.

Krugman is interesting because he appears to have given up on the amorality angle. This does not make him a traitor to the field, this simply acknowledges that the field is not impartial.

The questions that economists should answer in a political context is not "why is regulation bad/good?" but instead "What is the best policy given that my values are x,y,z?"

UK-AL• 11 years ago

It's not really amoral views. Take minimum wage. If they genuinely believe that minimum wage increases unemployment. Then they have taken the moral view that less unemployment, is better than better pay for the workers that stay in work. Most people would agree.

skwirl• 11 years ago

The problem is that there is no data that has shown this to be true for the types of minimum wage increases that are commonly considered. I think the actual moral view is that the government should not restrict the freedom of employers and employees to enter into whatever voluntary agreements they want, regardless of the outcome. Most people would disagree.

eli_gottlieb• 11 years ago

Then they have taken the moral view that less unemployment, is better than better pay for the workers that stay in work. Most people would agree.

How do both those alternatives play out when considered in terms of total take-home wages and the split of revenues between labor and capital?

furyg32• 11 years ago

But what is the rationale behind suggesting in minimum wage? It's not 'more pay'... it's something like reducing poverty, 'fairness', a moral belief that people should not suffer, etc.

The answer "Laws requiring employers to pay [higher wages, better pensions, healthcare coverage] increases unemployment because regulation has negative unintended consequences" a) is useless to policy makers and b) is a dismissal of the original moral position. It is thus a moral stance in itself. It says "I value freedom of the market more than I value your goals." I don't know if most economists have been manipulated into this role or are willing contributors, but the result is that they are being taken less seriously by society.

Economists are currently in the role of saying that the rising tide of a free market will eventually lift all boats, don't be distracted by the losers... but we seem to be reaching the limits of the effectiveness of this strategy and society seems to be uncomfortable with the casualties.

The question that generally needs answering is: "How can I reduce the negative consequences of free markets and what is the most effective way to compensate the losers?"

Retric• 11 years ago

It's important to note the difference between short term and long term effects.

Yes, unemployment might increase in the short term, but if cheap workers can't fill a need then people will look for alternatives like increased automation.

For a concrete example, machines can automatically dispense correct change, yet due to low wages this is fairly uncommon.

PS: To make this work we long term we need a minimum income, but that's a side issue.

netcan2• 11 years ago

For some context..

Adam Smith saw himself as a moral philosophy specialist. He spent way more time and effort on it that economics. He saw economics as an important tangent to moral philosophy.

His most important book in his eyes was "A Theory of Moral Sentiments" not "Wealth of nations."

calibraxis• 11 years ago

Transparently fake duality: "economic theory" vs "hating economics". When obviously there's many kinds of economic theories he decries. (Many of which are no doubt better at understanding/predicting than the ones he espouses, because his "theories" are required to serve certain sectors of power.)

netcan2• 11 years ago

I am almost certainly missing important nuance as an outsider. One thing I find interesting about US right wing politics (the US right, is I think a more uniquely American camp, then the left) is the way "rhetoric" is possible more accurate than policies.

Economically, the US right is very vocal and ideologically clear about its beliefs, economic liberalism. The politicians, pundits and supports cite fundamentals quite a lot. The efficiencies and incentives of free markets, with policies relating fairly directly to economic textbooks (of agreeable persuasions).

I don't want to stigmatize by calling it "economic ideology" as this implies an irrationality that I don't mean to imply, but ideology is more or less what I mean, sans value judgements. Its unfortunate that has become a bad word.

This writer is calling it "economistic" and means a strong expectation that the basic economic models like supply and demand are predictive of the real world effects. So, he expects the minimum wages to result in unemployment. This may be the real rationale for him, but I think the wider political weight of this is only there because it dovetails with other (US-sh) freedom concepts. IE, I think the idea that two people can decide on a wage contract without outside intervention has an appeal beyond just it's economic effects to many in that political camp.

In any case, my point is that this kind of theoretical grounding and economic fundamentalism (again sans-value judgment, I just mean they often tie policies to theoretical fundamentals) is something in recession in other places and political camps. The big examples are the bastard descendants of communism, arguably the most confidently touted and complete economic-political big-T Theory in history.

For example, take "labour-socialism" the big central-left camp in europe. Once upon a time, they believed in labour unionism a variant of socialism. Not just as a means to an end, but as a fundamentally grounded ideology. The way to make industry thrive, create prosperity for workers, etc. It still exists as a real-politic hodgepodge of policies and rhetoric roughly espousing old labour values, but not really labour unionism ideology. It's not comparable to the economic liberalism of the US right.

The Chinese communist party was once… communist. 5 year plans. means of production, the whole shebang.

The remaining left in Europe and elsewhere is very non theory based. The have strong a social and economic goals and values, but not really a big theory about how everything works best. There is no consistent, underlying economic theory like the US right has. It still holds a lot of socialist values and even

We get back to words like idealist or fundamentalist that have negative connotations, an interesting thing in itself. I think that's what this article is arguing for, essentially. Fundamentals say minimum wage increases yield unemployment. Trust in the fundamentals.

quadrangle• 11 years ago

FWIW, minimum wage could be shown to be basically positive and not cause all the unemployment people speak of and still be problematic. The issue with minimum wage isn't that it rejects supply-and-demand (that's simplistic thinking), it's that it is means-tested. Minimum wage doesn't directly help independent contractors or self-employed folks…

We need a Universal Basic Income and then to get rid of welfare and minimum wage and such.

But seeing as that won't happen soon, there's a very interesting positive aspect to minimum wage: People working for only social reasons such as volunteers or unpaid apprenticeships have a different psychosocial relationship to their situation than those who are paid an hourly wage. It's far better to be paid nothing than to be paid 10¢ an hour. Paid nothing means everyone acknowledges that you are volunteering your time, and it doesn't mean you are worthless. Paid 10¢ means you are worth 10¢. This has been studied by folks like Dan Ariely.

So, a minimum wage requires that people's time have a dignified enough value put on it, and anything less you must work with them on a volunteer, social basis.

I still think it has problems and UBI is better, but this issue of dignified pay should be appreciated.

bko• 11 years ago

Modern economics seems to be relying on more and more empirical data rather than theory. I think this is unfortunate. If you throw a pebble into the ocean, you cannot measure that this pebble increased the water level but no one would argue otherwise. No empirical data necessary and it would be silly to argue anything else. Similarly, artificially increasing the cost of hiring labor will decrease the quantity demanded.

Put another way, if one were to ask an economist how does the price artificially increasing the price of X affect the quantity demanded of that product, an economist would easily be able to answer without much hesitation or inquiry into the product X. I think a higher burden of proof should be placed on those claiming that artificially increasing the cost of labor doesn't affect quantity demanded. Since there are never any real controlled experiments in economics (states or municipalities all have their own history and other factors at play), you have to rely on theory.

Someone can agree for an increase in the minimum wage, but at least be honest about it being based on a moral trade-off between decreased employment opportunities for those whose labor is not above the current clearing rate and the benefit to others who would keep their jobs and enjoy higher pay.

gjm11• 11 years ago

It's true that economics is not amenable to experiments as reliably informative as those of, say, physics. And it's true that, all else being equal, the less reliable your experiments the less weight their results should get relative to theoretical predictions.

But it isn't only the experiments that are less reliable in economics than in physics. The theory is less reliable too. (For two reasons. Firstly, because it's so hard to do experiments whose results you can trust, the evidential basis for the theory is weaker. Secondly, because people are more complicated than, say, electrons.)

How does this cash out in the case of the minimum wage? We have some empirical evidence suggesting that minimum wage increases don't increase unemployment much if at all, but we can't trust it too much for the usual reasons (not enough data, too many confounding factors, small effect sizes). And we have some simple theoretical arguments suggesting that minimum wage increases should increase unemployment quite a bit (supply and demand from Econ 101), but we can't trust those much either because the real people involved in the labour market are not the perfectly rational perfectly selfish perfectly spherical agents with which Econ 101 is concerned. It simply isn't true that increasing cost always reduces demand: there are multiple mechanisms by which in particular circumstances it can do the reverse. It simply isn't true that the labour market behaves exactly like the market for, say, corn. People are complicated.

So I don't agree with what I take to be your argument about the minimum wage, namely that because we don't have great empirical evidence we must rely on theory and firmly believe that increasing the minimum wage will severely increase unemployment, and that this means that an honest supporter of minimum wage increases is obliged to accept that their preferred policy will increase unemployment substantially.

Incidentally, I'm not even so sure about that pebble thrown into the ocean. The sea-level rise is a tiny effect, after all, so we can't just ignore other tiny effects. For instance: a slight variation in the shape of the earth, with less mass on land and more on the ocean bottom; a tiny change in ocean currents, which might change erosion patterns and change the volume of the oceans one way or another; maybe some funny effect via disturbing local ecosystems and ultimately affecting human food supplies a little. These are all really small, but so is the obvious effect of throwing the pebble in.

For that matter, if you're on a boat when you throw the pebble in, the pebble goes from displacing its own mass of water (thank you, Archimedes) to displacing its own volume of water, which is smaller, so the immediate first-order effect is that the water level goes down.

metapattern• 11 years ago

The author suggests that Krugman has "changed the way he thinks". I would like to suggest instead that his perspective of the world around him may have changed.

For instance if he previously saw the liberals holding a disproportionate amount of power to sway public opinion, he may have attempted to re-balance the situation.

More recently, he may see conservatives using economic rhetoric to push their own agendas, and follows his previous pattern of re-balancing the playing field.

In both cases, he may be subtly promoting his own view, just in response to a changing audience.

Of course my opinion is just a projection of my own perspective.

Two ways of thinking about economics

Related Stories