A Little Perspective

As recently as a few months ago, doctors were held in high esteem and educated people believed that medicine could be useful. All that changed, of course, with the medical profession’s stunning failure to prevent or even predict the breakout of ebola in West Africa. Worse yet, many doctors to this very day cling to their old ways of thinking, writing prescriptions, setting broken bones, and performing surgery in bull-headed defiance of the urgent need to jettison everything we know about medical practice and start over from scratch.

Nobody, of course, writes such nonsense about medicine. Why, then, do so many write equivalent nonsense about economics?

Most economists failed to predict the 2008 financial crisis and ensuing recession for pretty much the same reason most doctors failed to predict the 2014 ebola epidemic — their attention was, quite reasonably, directed elsewhere. It’s easy to say in hindsight that if economists had paid more attention to the shadow banking system, they’d have seen what was coming. But attention is finite, and if economists had paid more attention to the shadow banking system, they’d have paid less attention to something else.

For a little perspective, have a look at this chart showing U.S.~per capita income in fixed (2005) dollars:


That little downward blip you see near the top is the recent crisis. The somewhat bigger downward blip in the 1930s is the Great Depression. The moral is that in the overall scheme of things, recessions don’t matter very much. At the trough of the Great Depression, people lived at a level of material comfort that would have seemed unimaginably luxurious to their grandparents. Today, while Paul Krugman continues to lament “the mess we’re in”, Americans at every income level live far better than Americans of, say, 1980. If you doubt that, you surely don’t remember what life was like in 1980. Here’s how to fix that: Pick a movie from 1980 — pretty much any movie will do — and count the “insurmountable” problems that the protagonist could have solved in an instant with the technology of 2014. Or reread any of the old posts on this page.

If you care about human well-being, recession-fighting is small potatoes. It’s that long-term upward trend that matters. And economists, fortunately, understand a lot about what it takes to nourish that trend — things like well-enforced property rights, the rule of law, free trade, sound money, limited regulation and low marginal tax rates. Even more fortunately, economists have managed, however imperfectly and with fits and starts, to impress that understanding on the minds of policymakers. As a result (and going back, at least, to the repeal of the Corn Laws), we’ve had better policies and greater prosperity.

To throw out all that hard-won knowledge because we failed to prevent a financial crisis would be like closing all the hospitals because doctors failed to prevent an epidemic.

Moreover, it’s entirely possible that some of the best policies for fighting recessions are inimical to long-term growth. It could easily follow that even if you knew exactly how to fight recessions, you might prefer not to.

It’s a very good thing that some economists are trying to understand recessions, and a very good thing that they’re accounting for the lessons of the past few years. It’s also a very good thing that most economists are working in the myriad of other areas where we’re capable of doing good. Another very good thing is historical perspective. The current so-called “mess” that economists have (partly) gotten us into is not just the most prosperous era in human history; it is prosperous beyond the wildest imaginings of your parents’ generation. And yes, economists helped get us here.

Print Friendly, PDF & Email
Share

74 Responses to “A Little Perspective”


  1. 1 1 James

    Steve, do you have a similar chart for real median individual (not household!) income?

  2. 2 2 jb

    I’d love to see the same plot but with a logarithmic y-axis, if it’s not too much trouble.

  3. 3 3 Andrew Whitby

    I agree with you that we shouldn’t ignore the long term trend. But I don’t think you’ve made the case that “economists helped get us here”.

    A full response here:
    http://andrewwhitby.com/2014/10/06/what-is-the-legacy-of-economics-so-far/

  4. 4 4 Joe

    Nice post. I hadn’t thought of it quite that way, but of course you’re completely right.

  5. 5 5 Steve Landsburg

    Andrew Whitby: Your post is excellent. I hope everyone who reads mine reads yours too.

  6. 6 6 Tjd

    Interesting post,

    Your analogy is off; bad economics resulting into bad policy (it happens) is akin to pharmaceutical companies selling bad drugs. The pharma industry can actually prove with the scientific method whether their drugs are bad or not and they still get it wrong at times.
    Economics is not scientific (cannot use the scientific method) so debating the virtue or value of economics is best done with beer or in an arm chair.

    Any economic theory that relies on ‘homo economicus’, ‘ceterus paribus’, ‘invisible hands’, ‘free markets’ (not even the stock market is a free market) is in essence relying on gross simplification and magical thinking.

    Regardless, I agree on the need for well-enforced property rights, the rule of law, and sound money. You don’t need economics to prove this, I think studying history is sufficient.

    I think free trade (which country practices this?, please don’t say the US), limited regulation and low marginal tax rates are still not proven.

    Finally, your thoughts on poverty in the modern world belong in a Dickens novel.

  7. 7 7 Steve Landsburg

    Tjd:

    bad economics resulting into bad policy (it happens) is akin to pharmaceutical companies selling bad drugs.

    And when a bad drug gets sold, do we typically hear people clamoring that everything we thought we knew about pharmaceuticals is hogwash?

  8. 8 8 Alan Wexelblat

    I’m curious what you make of the fact that the slope of that curve seems remarkably invariant to the things you list (well-enforced property rights, the rule of law, free trade, sound money, limited regulation and low marginal tax rates). If we look across the span of the 20th century I think we’d agree that there have been huge variations in how the US has handled each of those things. Yet if we pick segments of the curve in which any of these things varied greatly – say, marginal tax rates in the 70s vs the 90s – we find what seems to me to be very little change in the slope of the line.

    If that’s so, what’s the argument that these factors are at all relevant to the slope?

  9. 9 9 Satish

    Why are they mutually exclusive in the way you put it. For starters there is enough research to indicate that students who graduate during recessions face worse outcomes than those who graduate during normal times.So, at least for them there are long run implications. I don’t think a doctor would tell his patient with a brain clot in head to ignore it coz in the long run eating healthy and exercising are the only things that would matter.

  10. 10 10 Chris

    Andrew and Steve,

    There is a nice interplay there. But here is a question to both of you: Is the 20th century experiment in central planning an example of how economists’ ideas allowed us to get to our current level of prosperity? If not, why not? If the answer to that question is no because you consider it just a political or military conflict, then we need to rethink Marx’s place in intellectual history as well as how the calculation debate influenced ideas. But if there aren’t clear examples of economists’ ideas pushing us in the right direction, maybe we ought to think of them as preventing us from reversing course.

  11. 11 11 Steve Landsburg

    Alan Wexelblat: That’s a great question. The evidence I had in mind is all based on cross-country comparisons, as opposed to comparisons over time.

    Of course with cross-country comparisons, you need to worry about whether policies and growth rates are correlated not because one causes the other, but because they’re both caused by the same thing (culture, climate, etc.). There’s a lot of good work that attempts to control for this (e.g. Daron Acemoglu’s), and I find it pretty convincing.

    As to what you’re seeing on the US graph, my first thought is that, compared to what we see around the world, there hasn’t actually been very much policy variation in the US over time. But I wish I had a better answer for you.

  12. 12 12 Jeremy H.

    @James

    I believe the oldest consistent series with Median Individual Income only goes back to 1947: http://www.census.gov/hhes/www/income/data/historical/people/2013/p02.xls

    @jb

    You can create your own logarithmic graph here (along with tons of other historical data): http://www.measuringworth.com/usgdp/

  13. 13 13 Steve Landsburg

    Jeremy H: Thanks for those. I was about to reply to James saying that I think there’s no hope of getting median income data going back before about 1950, when you saved me the trouble.

  14. 14 14 Ken B

    @Steve #7
    “And when a bad drug gets sold, do we typically hear people clamoring that everything we thought we knew about pharmaceuticals is hogwash?”

    Actually, alas, sometimes we do. I hear endlessly about how anti-depressants are happy pills, have no effect, make things worse, mask the problem, are a non-solution to a non-problem.

    (Full disclosure: better half works in the pharma industry but not on anti-depressants.)

  15. 15 15 Charles G. Phillips

    Professor Landsburg,

    What is reasonable about economists having their attention directed elsewhere while the country almost committed financial suicide? What important events were occurring in 2008 that deflected your attention from the housing bubble, the dramatic increase in bank leverage and the invidious effects of Fed policy? Doctors are not being criticized for failing to anticipate the outbreak of Ebola, but surely this is a false equivalency for the failure of most economists to properly observe the dangers contained in the data that they review as a matter of course. A better comparison would be between distracted economists and the doctors at the Dallas hospital who, despite all the evidence suggesting a need for caution, sent the Ebola patient home after his first visit to the emergency room. When doctors make mistakes they are sued–quite often successfully–and subjected to penalties that include the loss of the right to practice their profession. Tenured economists’s worries involve a different category of concern, such as when to replace the Volvo station wagon and when to start coloring their beards. Regards, Chas Phillips

  16. 16 16 Ken Arromdee

    I would guess that, say, $7500 in 2005 is probably a starvation wage. Since this graph shows that income is below $7500 in 2005 dollars prior to about 1935, I conclude that everyone before 1935 must have starved to death, and accordingly we are not here.

    Are you really sure that 1) the chart is in fixed dollars, and 2) the chart includes as income such things as people growing/raising food for their own use?

  17. 17 17 nobody.really

    Nobody, of course, writes such nonsense about medicine.

    Landsburg reads my medical blog? I’m flattered!

    And when a bad drug gets sold, do we typically hear people clamoring that everything we thought we knew about pharmaceuticals is hogwash?

    As I said on my blog, we might — if they’re still under the influence of the drugs.

    I was about to reply to James saying that I think there’s no hope of getting median income data going back before about 1950, when you saved me the trouble.

    I would guess that, say, $7500 in 2005 is probably a starvation wage. Since this graph shows that income is below $7500 in 2005 dollars prior to about 1935, I conclude that everyone before 1935 must have starved to death, and accordingly we are not here.

    I also was about to reply to James, when Ken Arromdee documented than none of us are here, thereby saving me the trouble. Thanks, Ken!

  18. 18 18 CC

    “At the trough of the Great Depression, people lived at a level of material comfort that would have seemed unimaginably luxurious to their grandparents.”

    You made a similar argument in one of your books (I think it was MSISS — awesome book btw). You said something like, “in the Great Drepression, people’s standard of living was rolled back to what it was just several years ago.”

    The problem I see is that what makes recessions bad is not that average incomes get scaled back slightly, but rather that a few people (maybe 5%) completely lose their jobs and can’t find new ones. After all, if the president said, “We’re about to go through a recession, and everyone’s income is going to shrink by 2% next year, and then recover over the following year,” this really wouldn’t sound too bad. But if we knew that instead most people would keep their current income, a bunch of people would lose their jobs and be out of the labor market for a long time (or forced to retire) and the rest of us would live in fear that we’re next to become unemployed, suddenly things look a little scary.

    So yeah, long term growth is really awesome, but I think you downplay the pain of recessions.

  19. 19 19 Steve Landsburg

    CC: One of the costs of a recession is that it exposes people to the risk of being hit a lot harder than the average person. You are absolutely right that to calculate the cost of a recession, we should include that risk as a cost — valuing it at the amount that we estimate people would be willing to pay to avoid that risk. We can form that estimate by looking at data on how much people are willing to pay to avoid other sorts of financial risk — for example, how much average income they’re willing to give up for the privilege of working on salary rather than on commission.

    I agree with you that I have not addressed this in the post (or in the book). But I claim that if you *do* address it, you still find that the costs of a recession are pretty small. I should blog about this sometime.

  20. 20 20 CC

    SL: Very cool! I look forward to a post about that. (As well as another book, along with a post about the relationship between different infinities and the power of proof systems…)

  21. 21 21 Harold

    “Pick a movie from 1980 — pretty much any movie will do — and count the “insurmountable” problems that the protagonist could have solved in an instant with the technology of 2014”

    I tried this out. The highest grossing film of 1980 was The Empire Strikes Back – not much help there with 2014 technology. Sci-fi aside, a pretty good proportion of films from 1980 were about how the guy gets the girl or vice versa. Technology would not have helped any of these protagonists.

    True, a mobile phone or internet connection might have been pretty useful in The Shining, but it would be easy to arrange for these to be absent from the isolated, out of season hotel.

    Not really sure how Airplane would be different with today’s technology. Surely, things have stayed pretty similar, but I think the autopilots are not inflatable today (and don’t call me Shirley).

    The Gods Must Be Crazy would stay pretty much the same. Coke bottle might have changed shape.

  22. 22 22 Harold

    #14 From ScientificAmerican:
    “A controversial article just published in the prestigious Journal of the American Medical Association concluded that antidepressants are no more effective than placebos for most depressed patients. Jay Fournier and his colleagues at the University of Pennsylvania aggregated individual patient data from six high-quality clinical trials and found that the superiority of antidepressants over placebo is clinically significant only for patients who are very severely depressed. For patients with mild, moderate, and even severe depression, placebos work nearly as well as antidepressants.”

    It seems that antidepressants are just a way of prescribing placebos so the doctors are happy with is and the patient benefits from the placebo effect. A grand result all round, as it seems a shame not to take advantage of such a powerful effect.

  23. 23 23 Ken Arromdee

    The census data at https://www.census.gov/hhes/www/income/data/historical/people/ for per-capita income in constant dollars, all races (table P-1) shows 28829 in 2005 and 15026 in 1967, where it ends, which has a slope reasonably similar to this graph. However, *median* income (table P-5) shows 35228 in 2013, 33862 in 1967, and 24596 in 1953, the first year listed. That’s an increase in 43%–which is nowhere near the increase of 200% implied by the graph or by extrapolating the mean income figures. This indicates that the graph is heavily distorted by a few people making a lot of money.

    It also seems to be distorted by birth rates (the median figures exclude babies and children. I also suspect that the early years fail to count people growing food for their own use as having income.

  24. 24 24 iceman

    Fair point. Of course in a market economy, “a few people making a lot of money” = a few people creating stuff that we like, which gets back to all the luxuries like iPhones we enjoy today even on a sub-median income. Admittedly unclear how mortgage-backed securities fit into this story.

  25. 25 25 nobody.really

    The census data at https://www.census.gov/hhes/www/income/data/historical/people/ for per-capita income in constant dollars, all races (table P-1) shows 28829 in 2005 and 15026 in 1967, where it ends, which has a slope reasonably similar to this graph. However, *median* income (table P-5) shows 35228 in 2013, 33862 in 1967, and 24596 in 1953, the first year listed. That’s an increase in 43%–which is nowhere near the increase of 200% implied by the graph or by extrapolating the mean income figures. This indicates that the graph is heavily distorted by a few people making a lot of money.

    Let me see if I’m following this:

    For the 60-year period 1953-2013, Ken’s census data shows that median per capita income increases from $24,596 to $35,228 – an annual growth rate of 0.6%.

    Meanwhile, Landsburg’s graph shows that average US per capital income grew from roughly $13000 to roughly $45000 – an annual growth rate of roughly 2.1%, or 3.5x faster than the median, compounded for 60 yrs.

    Even if the data are not comparable, would the growth rates be comparable?

    Elsewhere, I read that the distribution of income growth during economic expansions skews heavily to the top 10% of the income distribution. To be sure, the top 10% own a lot of stock, so I’d expect some skewing. But the chart shows that from 1948-53, the top 10% garnered 20% of the growth. But by 2009-12, the top 10% garner 118% of the growth. (That is, the bottom 90% experienced a net loss in income.)

  26. 26 26 Steve Landsburg

    nobody.really (and several others): I am highly skeptical of the data on median income, because we’ve just been through a few decades in which the *unmeasured* components of income have increased dramatically, in ways that are probably far more equally distributed than the measured components. What’s the value of having 500 TV channels instead of 4? What’s the value of having all the world’s knowledge at your fingertips? These things are very large, and by rights ought to be counted as part of the rise in per capita income, though they’re not. Throw them in and you’re likely to see the median behaving a whole lot more like the mean.

  27. 27 27 Bruce

    Hi Steve,
    I can show you one economist that other well know economists thought was crazy (or just plain wrong). Peter Shiff.
    Shiff was 100% right on.

    See below video’s. If someone would have listened things may be better today.

    Bruce

    https://www.youtube.com/watch?

    v=S09jj7uagk0https://www.youtube.com/watch?v=1G0tfb8ZefA

    https://www.youtube.com/watch?v=s_z5dZbgTDM

  28. 28 28 Steve Landsburg

    Charles G Phillips:

    What important events were occurring in 2008 that deflected your attention from the housing bubble, the dramatic increase in bank leverage and the invidious effects of Fed policy

    This makes me wonder if you bothered to read my post before responding. Our attention was not “deflected” from the housing bubble, it was, in a great many cases, directed elsewhere to begin with. Here is a sampling of the topics my colleagues here at Rochester have been working on these past few years:

    The effects of trade liberalization, rent seeking and capital accumulation, the consequences of political competition, how sticky wages can affect hiring, measuring quality improvements in consumer durables, testing alternative theories of labor demand, the effect of smoking in pregnancy, the effect of breastfeeding on child development, general techniques for distinguishing between correlation and causation, the interaction between neighborhood patterns and school quality, testing the “broken windows” theory of criminal behavior, measuring the returns to schooling, optimal taxation in a world of income inequality, how labor market uncertainty affects household investment decisions, identifying the characteristics of observed random processes, the outcomes of bargaining, the consequences of various voting systems, the historical profitability of slavery, the consequences of various auction rules, the effects of trade on wages, the effectiveness of school suspension policies, measuring the value of a good teacher, understanding racial differences in school discipline, the relation between substance abuse and cognitive skills, the matching process between firms and workers, the use of quantum technology in strategic situations, the right way to think about optimization when memory is imperfect, the determination of city sizes, the impact of large devaluations on import markets, flows in and out of the labor market by workers in marginal jobs, the dispersion of prices and quality in offshoring markets, layoff costs, the relation between capital and employment at the firm level, how privately informed traders should negotiate, segregation in residential neighborhoods, how group size affects altruistic behavior, the optimal level of sovereign debt, austerity policies as constrained insurance contracts, the Fed’s influence on mortgage rates, the fair division of a fixed supply among a growing population, whether it’s possible in principle for everyone to benefit from growth, rules for adjudicating conflicting claims, the consequences of various rules for dividing the assets of bankrupt estates, how state and local governments cope with business cycles, testing for hidden asymmetries in strategic interactions, etc., etc., etc.

    So when you refer to “the failure of most economists to properly observe the dangers contained in the data that they review as a matter of course“, you betray a truly extraordinary ignorance about what most economists actually do. Why would economists studying school suspensions or allocation of bankrupt estates or altruistic behavior in groups or quits and layoffs be reviewing data on the shadow banking system as a matter of course?

    You could equally well say that all doctors are at fault for failing to predict the ebola outbreak, because of course all doctors study African epidemiology as a matter of course. But of course then you’d sound like an idiot.

  29. 29 29 Harold

    #26. Once you start to measure non-financial indicators it sounds like you are talking more about well-being than wealth. Whilst it is great to have 400 TV channels, it is not so great that your high paying job takes you 2000 miles from your ailing parents. Talking about well-being is fine, but a different kettle of fish.

  30. 30 30 Advo

    I think that the drop in esteem that economists suffered is well deserved.
    They had forgotten the lessons learned during the time prior to the Great Depression. Notably:

    If you give the banks too much rope, they WILL go and hang themselves.

    That, in a nutshell, is what happened. It’s what ALWAYS happens if you don’t impose stifling regulation on a competitive financial sector. There have been some periods of time in some countries without stifling regulation and without cyclical banking crises, but in those cases there was either a highly cartelized banking structure and/or unlimited personal liability for management and shareholders. A dynamic, competitive banking always evolves towards instability and meltdown because a business model that is 95% safe (meaning it’s safe unless there’s a credit crunch) is just way more profitable than a 100% safe business model.
    Plus, the 95% safe business model is actually 100% safe as long as only a relatively small part of the financial industry is doing it.
    Unfortunately, the risk-takers grow faster than the conservatives, end up dominating the industry, and eventually blow up themselves and a good part of the economy.

    Reinhard and Rogoff cite a handful of cases where financial de-regulation did not lead to highly destructive banking crises. The absence of disaster in the financial sector is the exception, not the rule.

  31. 31 31 Charles G. Phillips

    Professor Landsburg,

    Was it my comment about when to replace the Volvo station that got you? I re-read my post: it was snide, for which I apologize. The validity of your main point (it is the upward trend, not the downward blips that matter) stands on its own. The analogy you employed to set up your main point is mis-guided, for reasons I thought I made clear in my original response. No one is suggesting that we start from scratch in either medical practice or economics, but the notion that the failure of doctors to anticipate the outbreak of a virus in West Africa is comparable to the the failure of most U.S. economists (those paying careful attention and those otherwise engaged) to miss the largest financial catastrophe of their lifetimes strains credulity.

    You noted that I expressed “a truly extraordinary ignorance about what most economists actually do” after listing the subjects that have been occupying the Rochester economics faculty in great detail. You are probably correct. But I do not think that the following questions unreasonable: Does anyone on the faculty take an interest in Fed policy? Capital markets? Regulatory capture? The international banking system? Do faculty members read the Wall Street Journal, Economist, Financial Times, James Grant? Do faculty members talk amongst themselves about important topical matters, notwithstanding the heavy burdens of their research? I did not, in my response, state that economists should be criticized for not seeing the train wreck coming, but I did mean to suggest that, once you chose to put forward an excuse for this failure, you chose an incomplete and self-reverential one.

    Here is what I think happened. You had an interesting and important point to make about the seemingly small impact that recessions have had over the long term, and you had been thinking about Ebola (who hasn’t). You decided to mash the two together in what you thought would be a provocative way to set up your argument. Your analogy, unfortunately, was badly flawed and your defense of the failure of mosts economists to anticipate the financial crisis was and remains weak. My post annoyed you (maybe it was the comment about coloring beards) and you responded in kind. I don’t blame you after re-reading my post. Most of what I have learned about economics I have learned from you; it pains me that you think I did not read your post before responding and that you find me incredibly ignorant about what most economists do.

    Respectfully submitted, Chas Phillips

  32. 32 32 Daniel

    @Landsburg,

    I’m surprised your making this argument based on the US, which during the time period you cite has a consistent pattern of fighting recessions when they occur. It seems that you draw the conclusion that the slope of this line would be unchanged if instead the federal reserve and the federal government chose to do nothing. I have to ask myself given your rationality if you have some sort of reality distorting device that allows you to turn off key policies throughout the last 100 years? If so, why aren’t you sharing with the rest of the econ profession? We could solve a lot of arguments.

    I’m also surprised that you think anyone is interested in throwing out the lessons that economics has taught us. Who exactly is this person? I imagine you think it’s Krugman, but I’d like to see some quotes in which Krugman asks us to throwout the conventional wisdoms of economics, so that I can respond directly. Also, okay, so what if we are better off than we ever were before? Does that mean we shouldn’t strive for optimality? I guess my question is, what exactly are you trying to say here?

  33. 33 33 Harold

    After reading Advo’s comment, that might be a point. You say “the medical profession’s stunning failure to prevent or even predict the breakout of ebola in West Africa.” If you had asked doctors before the recent ebola outbreak what their major concerns were, epidemics would have been high on their lists. The doctors are trying to get us to take it more seriously and have been warning the next serious global epidemic is overdue. They freely admit that they cannot say exactly where or what disease will be the problem. If you had asked economists before the crash, that might not have been high on their lists.

  34. 34 34 Steve Landsburg

    Harold: Wealth measures the value of the resources at your command, i.e. it measures the limits of the consumption available to you. If you’re able to consume more, you’re wealthier.

    Think of it this way. Here are two histories of the past 40 years:

    In History A, Cable TV was invented. A package costs $10,000. Your income rose by $10,000. You use that extra income to buy a cable TV package.

    In History B, Cable TV was invented. A package costs $100. Your income rose by $1000. You use $100 of that extra income to buy a cable TV package.

    I hope you’ll agree that by any sensible measure, you are $900 wealthier in History B than in History A. But a measure of income that excludes the value of cable TV will say you’re $9000 wealthier in History A than in History B. So such measures — i.e. the measures people are citing in this thread — cannot be correct.

  35. 35 35 nobody.really

    I hope you’ll agree that by any sensible measure, you are $900 wealthier in History B than in History A. But a measure of income that excludes the value of cable TV will say you’re $9000 wealthier in History A than in History B. So such measures — i.e. the measures people are citing in this thread — cannot be correct.

    Such a rich vein.

    True, to the extent that (consumable) wealth data fails to reflect non-quantified benefits (and costs), it may lead to unjustified conclusions. This would be true of the measures of median income – and of average income. Indeed, conceptually the absolute difference between median and average would not change; the graph of median and average wealth would remain the same, only shifted upward by the value of a Cable TV package.

    That said, Landburg notes that the relative magnitude of that difference would change. Where the average number used to be X% larger than the median, now it would be < X%. And the larger the growth in non-quantified benefits, the smaller the relative distance between average and median.

  36. 36 36 nobody.really

    Wealth measures the value of the resources at your command, i.e. it measures the limits of the consumption available to you. If you’re able to consume more, you’re wealthier.

    Indeed. And many people are motivated to increase their ability to consume; that’s probably the theory that underlies your paycheck. Bear this in mind when people discuss the damaging effects of marginal taxes.

    According to this theory, people are motivated by how much after-tax ability to consume they gain for each additional hour worked. And this would remain true whether the marginal tax rate is 9% or 90%.

    The distinctions between average and median (whether or not you also include unmeasured components) should tell you that somebody is raking in vast sums relative what people used to earn. Yet people in the past found themselves motivated to work hard and produce nonetheless. To return to my favorite example, the fact that Johann Bach’s income was less than Justin Beiber’s did not seem to impair his output. I expect we could tax Beiber at 90% and he’d still end up with a greater financial incentive to work an extra hour than Bach did — or anyone else on this blog.

    So if you care about incentives, quit obsessing over marginal tax rates. Keep your eye on the ball: the stuff that’s left over after the taxes come out. That’s where the incentives lie — and for some segments of society, that stuff has been growing like crazy.

    (Admittedly, I’m putting aside the backwards-bending labor curve for the sake of this discussion….)

  37. 37 37 Floccina

    For a little perspective, have a look at this chart showing U.S.~per capita income in fixed (2005) dollars:

    That chart is why I loved it when shortly after the crash in 2008 some leftists would say: See where your free market has gotten us.
    I would say yes I do see where our free market has gotten us, despite the melt down no one is starving compare with North Korea where there is almost no free market. Consider that and then be very careful what free market stuff you jettison!

  38. 38 38 Floccina

    One other things is that all the recessions that most economists saw coming did not occur.

  39. 39 39 David Wallin

    Advo r/29
    I find the observation about banks and the great depression interesting. The banks that failed in the GD were mostly banks operated in rural areas in states that did not allow branch banking. States that allowed branch banking did much better. So did Canada, where branch banking was allowed and no banks failed.

  40. 40 40 Dan

    “So such measures — i.e. the measures people are citing in this thread — cannot be correct.”

    Wouldn’t that be true of the average income in the chart and its variation over time?

    I am always confused of discussions of average increases over long term. Is there any reason to care about the long-term or the average? It’s not like there is a representative person who lives 200 years and earns the average income. I find that there is an instant acceptance of the idea that technology or globalization is always good. They come with increases in average benefits and significant distributional costs. There is tendency to overstate the former and minimize the latter.

  41. 41 41 Dan

    nobody.really

    Excellent example with Bach and Bieber. I always thought of taxes and technology as substitutes. They both affect the marginal income. You can think of innovations in technology as a decrease in marginal taxes. People working in 1814 faced little marginal taxes but from the perspective of someone working 2014 faced high technology taxes.

    It gets a bit complicated as innovations themselves will depend on marginal tax rates, but the example is very good nonetheless.

  42. 42 42 iceman

    Daniel – “so what if we are better off than we ever were before? Does that mean we shouldn’t strive for optimality?”

    Of course the issue is the extent to which one’s view of optimality involves weighting SR vs. LR. Politicians are particularly incentivized to overweight the former, which if unchecked makes your “so what” a potentially big deal (especially for a utilitarian?). And certainly we’ve fought recessions in different ways e.g. early 20s and 80s vs. 30s and current.

    nobody 34-5: Again what struck me as interesting about avg vs. median is that “a few people making a lot” also reflects at least some of the value the rest of us get from their creations in a way median does not. Arguably consumer surplus makes the effect even stronger than either measure would convey.
    This seems consistent with the well-established principle that the important incentive effects happen at the margin. You insist that incremental work effort is not dependent upon how fully it is rewarded? Comparing different *levels* of income does not address this point, and you can’t simply set aside the opposing effect on the relative value of leisure. E.g. we might expect Bach to work harder (thankfully) than Bieber based solely on the income effect. This does not imply that Bieber does not work less than *he* would have if he keeps less (as much as we might wish it were so). Of course Bach might also be a uniquely inspired example.

    Dan – if “innovations themselves will depend on marginal tax rates”, to me this refutes the whole example. I don’t understand what is meant by “technology taxes” (since the only upper limit on them would appear to be our imaginations), other than this is the LR growth effect we want to promote.

  43. 43 43 nobody.really

    “a few people making a lot” also reflects at least some of the value the rest of us get from their creations in a way median does not. Arguably consumer surplus makes the effect even stronger than either measure would convey.

    Not sure what this means. Yes, consumers benefit from consumer surplus. Is there some reason to think that a world with larger wealth disparities generates more consumer surplus than a world with equal productivity but less wealth disparities?

    You insist that incremental work effort is not dependent upon how fully it is rewarded?

    No, I insist that (at least as I understand classical theory) effort is influenced by marginal ability to consume. Thus, I think I would be equally motivated by an incremental opportunity to earn $1000/hr — whether that was $1000 free of all taxation or $10,000 taxed at 90%. And because we live in a world in which there are ever more people earning ever higher incomes, we may be able to tax them at ever higher rates without diminishing effort (relative to a world in which people earned less and were taxed at a lower rate).

    I don’t understand what is meant by “technology taxes”

    This is another way of saying that people are motivated by the net marginal ability to consume. Net of what? Well, as I note, net of taxes. But as Dan notes, net of other costs. And in this sense, technology is akin to a negative tax: it reduced the costs that must be subtracted from gross income.

    This underscores the arbitrariness of obsessing over marginal tax rates: Sure, lower marginal rates reduce costs, all else being equal. But all else is rarely equal. Justin Beiber can earn more than Johann Bach because Beiber gets the benefit of the (negative) technology tax. Thus we could impose a 90% tax on Beiber’s income, and he’d still likely out-earn Bach.

    Now, if you could quantify a link between marginal tax rates and improved technological environment – “For each additional X% you pay in taxes, you generate innovations (e.g., improved roads, free-trade treaties, the internet) that improve social efficiency by an average of Y%” – then you’d have a real thesis on your hands, and might have something we could try to optimize.

  44. 44 44 Daniel

    iceman- “Of course the issue is the extent to which one’s view of optimality involves weighting SR vs. LR. Politicians are particularly incentivized to overweight the former”

    Except when short-term pain becomes long term pain as is the case when we have millions of discouraged workers whose skills are becoming rusty and for whom the longer they stay out of work, the higher the probability they become persistently unemployed. 20’s and 80’s were also completely different kinds of recessions between 30’s and now. The 80’s was a forced recession to put inflation expectations back into a reasonable territory. Same with the 20’s, a forced recession by the federal reserve (which they severely overshot and caused deflation). This is not at all related to our current situation, where the pedal is to the metal on monetary policy and we’re having persistently low inflation accompanied by high numbers of long-term unemployed. I’m just not sure where the confusion is coming in.

  45. 45 45 Harold

    OK, I see your point, I think, but there still seems something wrong with it.

    “Wealth measures the value of the resources at your command”

    How do we measure the value of something? Usually it is by what is paid for it. Is it possible that the value of the cable TV is different in each universe, even if the product is the same?

    Twenty years ago I was prepared to pay $1500 for a 286. The value of that computer was $1500. Today I would not be prepared to pay anything for it. Twenty years ago, the Met. office was prepared to pay several millions for a computer with the processing power of my current computer. Do I now say that the value of my computer is several millions?

    In your history A I would not choose the cable package, and I would be $9000 richer than in history B. I would have a new car instead of cable TV. Given the choice between the two histories I would choose A, even though you claim I am richer in history B.

    If you are saying that I get more out of my current computer than I got out of my old one, that may be true, but it does not mean it has a higher value. My valuation is simply different.

  46. 46 46 nobody.really

    How do we measure the value of something? Usually it is by what is paid for it.

    I agree: that is typically how we measure such things. But rather than indicating a problem with Landsburg’s analysis, this indicates a problem with how we measure things. (Or rather, a problem with our incomplete understanding of such measurements.)

    Price reflects the minimum that buys value something, and the maximum that sellers do. Yet graphs of supply and demand reveal consumer and producer surpluses. Because we don’t directly observe the surpluses, but do observe the sales price, we focus on the data that’s easy to identify.

    Twenty years ago I was prepared to pay $1500 for a 286. The value of that computer was $1500. Today I would not be prepared to pay anything for it. Twenty years ago, the Met. office was prepared to pay several millions for a computer with the processing power of my current computer. Do I now say that the value of my computer is several millions?

    If we assume your tastes and preferences have changed over 20 years, than you’re right, the analogy does not hold. (Alternatively, we could say that the two people — you 20 years ago, and you today — are not really the same person.)

    But I don’t think you mean to say that your demand for computing power has changed. Rather, you’re pointing to the fact that over the past 20 years the supply of competitive alternatives has changed, which has altered the quantity demanded of computing power. Conceptually, both you and the Met may have the same demand for computing power as you did 20 years ago. That is, if a disaster struck and we were thrown back into a world of 1990s technology where you had the choice of buying a 286 for $1500 or having no computing power at all (i.e., if the supply curve shifted left), you might once again be willing to pay that price (i.e., it would cross your demand curve at a point indicating a higher price and a lower quantity). Today you don’t pay that price, not because you have no more use for computing power, but because you have better alternatives at lower prices.

  47. 47 47 Al V.

    Regarding CC’s point about the distinction between the impact of the Depression on “average” Americans, and the impact on people who were disparately affected, I think my family is an excellent example.

    My father’s parents were university professors. The Depression had little impact on their incomes, and the fall in property values allowed them to buy a farm as vacation property.

    My mother’s father managed a grocery store that went out of business. He was unable to find a new job, and the stress ultimately killed him at 50.

    On average, my grandparents experienced an approximately 30% drop in income. But that’s probably not the measure we should use.

  48. 48 48 iceman

    Nobody – Let’s try it this way: presuming we all agree Steve Jobs made life better for a lot of us, this is partly reflected in the amount of money he made in the process (what we were willing to pay – at the margin – while still not capturing our surplus), but not in median income.

    To me you frame the issue backwards – a particular wealth dispersion does not generate consumer surplus, both result from different “productivity events”. It’s meaningless to put these into “all else equal”.

    And surely opportunities that pay $1,000/hr versus $10,000/hr pre-tax are not the same opportunities. How much we choose to tax the latter will impact incentives. And the wealthier people are the more the leisure substitution effect kicks in. There must be reams of work that have been done on marginal incentive effects. I doubt my alma mater had it that wrong.

    Daniel – I know many believe things like liquidity traps and hysteresis effects can suggest the SR/LR tradeoff doesn’t apply, at least under circumstances that may have applied in two instances in the graph at hand? As you cite, sometimes policymakers have also made things worse. I don’t believe anyone is saying we shouldn’t try to learn what we can from “this time is different”, just that it’s far from the only thing that matters.

  49. 49 49 Al V.

    Re. the “*unmeasured* components of income”, a good comparison here is me and my son. When I was 24, I had a television (with cable), a VCR (renting tapes at a video store), a stereo (and 100 or so albums), mailed letters, and paid bills via mail. My son has a laptop and an internet connection. I paid more for the stereo alone than my son paid for the laptop. In constant dollars, I probably paid three times as much for the stereo as he did for the laptop.

    If I wanted to watch a movie, I could choose from any movie stocked at the local video store. If I wanted to listen to music, I had 100 albums. My son has Netflix, Hulu, and Spotify.

  50. 50 50 nobody.really

    And surely opportunities that pay $1,000/hr versus $10,000/hr pre-tax are not the same opportunities.

    Not the same – with respect to what variable? (dt? dx?)

    For example, you might choose to invest $X to buy a small farm in England that generates revenues of $1000. Or you might choose to invest this sum moving to the New World where land is cheaper, and farm a bigger stead that generates $10,000. But the Crown then imposes the Stamp Act to pay for the troops that defend your farm during the French and Indian Wars, and this Act costs you $9,000, so you net $1,000. Are these two things “the same opportunity”? With respect to the opportunity to earn a return on $X, yes. In other respects, no.

    One classic way the sums get fudged is to fail to account for risk. In this sense, two events that each produce $1000 may be completely different, because one involved more risk than the other. Perhaps Landsburg’s thesis is that economists were taken by surprise by the market crash for completely rational reasons: the returns of focusing on growth swamp the returns on guarding against loss. If you can learn to score a touchdown with each play, why invest in a defensive line?

    Lax financial regulations arguably generate both 1) increased productivity through risk-taking and 2) crashes. It is not self-evident that one effect predominates – and Landsburg hints that the former offsets that latter. After all, Reagan presided over the Savings & Loan debacle (costing taxpayers something like $500 million?), and over explosive economic growth. If we could only provide mechanisms to compensate the innocent parties who are harmed by the crashes, the optimal strategy may be to simply let the good times roll and crash, roll and crash.

    And the wealthier people are the more the leisure substitution effect kicks in.

    Yeah, that’s the backwards-bending labor curve. And one remedy for that curve is – lump sum taxation! That is, if you can keep people from becoming rich in the first place, then they don’t get to the point where they will forsake opportunities to earn $1000/hr merely to get an extra hour of leisure. But the tax can’t be put on any marginal event, or else it would distort things at the margin. So we need something like an unanticipated Community Chest card that says “Street Repair: pay $XXX for each house and $XXX for each hotel….”

  51. 51 51 Jack PQ

    @31 Charles Phillips,
    I encourage you to read both Paul Krugman’s “How did economists get it so wrong” and John Cochrane’s rejoindre “How did Paul Krugman get it so wrong” to get an answer to your question.

    Other points to add:
    (1) By definition, a crisis cannot be anticipated. If it is anticipated, the adverse news are incorporated gradually rather than suddenly, and there is no crisis. So “anticipating crises” is a straw man.
    (2) After event X, it seems obvious more people should have spent more time working on X. Hindsight is 20/20. Back in the 2000s there was no obvious smoking gun. Economists collectively try to figure out the best questions to investigate. Maybe they get it wrong in an ex post sense. But nobody has a crystal ball.
    (3) One point where I think you are right is that the link between finance and macroeconomics was perhaps underestimated, and since 2008 there is a lot of work to better understand this area.
    (4) Most economists don’t read the financial news, nor do they try to forecast, or beat the stock market. They do scholarly research on large and persistent issues, not ephemeral blips on the radar (as the news is).

  52. 52 52 Harold

    “By definition, a crisis cannot be anticipated.” We anticipate potential crises all the time, and adapt our behaviour to avoid them. The ones that evolve into actual crises are the ones we did not act on.

    Anticipating a specific crisis is not the same as anticipating that a crisis is likely to happen. Back to the doctors, they anticipate that there will be a pandemic flu sometime. They cannot say exactly when. One could hypthetically predict that with a certain type of regulation a financial crisis was inevitable, but you could not predict exactly when it would happen.

  53. 53 53 CC

    nobody.really: The correct way to measure the deadweight loss from high marginal tax rates is (usually) to just look at the substitution effect, not the income effect. Garett Jones explains it here: http://econlog.econlib.org/archives/2012/10/prescott_hoisti.html
    and Mankiw explained it a while back here:
    http://gregmankiw.blogspot.com/2007/04/do-income-effects-matter-for-tax-policy.html

    These posts were eye-opening for me. I never understood this before, even after multiple econ classes.

  54. 54 54 Ted Levy

    “Here’s how to fix that: Pick a movie from 1980 — pretty much any movie will do — and count the “insurmountable” problems that the protagonist could have solved in an instant with the technology of 2014”

    This is exactly why some stories don’t translate well from novel to movies if sufficient time has passed. Example: Dagny Taggart in NYC desperately needs to get hold of Quentin McDaniels in Colorado. SO SHE GETS ON A TRAIN to travel to him. When the train breaks down, she rents a small plane she fortunately knows how to fly. And all the kids in the theater are thinking, “Why doesn’t she just text him…or use FaceTime or Google+ or Skype…?”

  55. 55 55 iceman

    50 – I find your example too contrived to be useful, as a “choice” between two jobs paying $1,000 or $10,000 pre-tax would reflect different skill sets, and/or risk as you suggest. E.g. I suspect most people who endured the hardship of moving to a new world did so in the hope of doing materially better (I’ve heard those tax acts created quite the little uprising). Or else you’re suggesting we can tax a given opportunity however we wish and the gross simply adjusts?

    “if you can keep people from becoming rich in the first place” – I can only presume you’re being facetious (better to punish *every* marginal effort past and present in one shot?). Maybe you can fool all the people one time…see comment on colonial tax acts above.

    I do agree the post seems to support the idea that focusing on the right growth policies increases the pie sufficiently that we could help those who lose out in the “crashes” (where we felt this was “fair” and could properly identify them) and thereby ensure that everyone is better off in the LR (certainly all future people). In fact we kind of already do this in a variety of ways, which don’t even account for the prevalence of smartphones.

  56. 56 56 Daniel

    @ Iceman,

    “I know many believe things like liquidity traps and hysteresis effects can suggest the SR/LR tradeoff doesn’t apply, at least under circumstances that may have applied in two instances in the graph at hand? As you cite, sometimes policymakers have also made things worse. I don’t believe anyone is saying we shouldn’t try to learn what we can from “this time is different”, just that it’s far from the only thing that matters”

    Umm, we screwed up in the 1920’s well before Friedman’s and before a lot of New Keynesian models were developed. So far, the monetarist and New Keynesian tool-sets have worked relatively well. The times that have not worked recently have been when we’ve ignored these toolsets and listened to “reasonable people”.

    If one needs an example of what not fighting recessions with known policy tools looks like we need only look to Europe. It’s a perfect demonstration of what happens when we say they are more important things than dealing with the most pressing issue at hand. SR issues start to become LR issues.

    And can I say more generally, that Professor Landsburg has particularly bad timing to be trying to make the argument that fighting recessions should not be one of our main priorities given recent events in Europe!

  57. 57 57 Daniel

    “just that it’s far from the only thing that matters.”

    No one has ever made the argument that fighting recessions is the only thing that matters. This is a strawman. The point is, when your ship has a big whopping hole in it, you patch the hole, and then worry about the speed of your sails. Landsburg is essentially saying, look how fast our ship is, patching holes really didn’t make that big of a difference in the distance we travelled, without acknowledging that we’ve been patching holes as we go along the way.

  58. 58 58 nobody.really

    So … how ’bout that Nobel Prize?

  59. 59 59 iceman

    I did say it was *far* from the only thing.
    We agree that you have much more faith in policymakers and fine-tuning than I do. I see SR becomes LR in the sense that each time we decide we must do whatever it takes to patch a hole we weaken the fabric leading to another bigger hole.
    I think the consensus still believed in an exploitable Phillips Curve long past the 20s, which as you note required an intentional recession to fix. But I’m sure we could never repeat such a mistake. Like today we know when a big stimulus doesn’t work that proves it should have been much bigger. At least one thing we should know pretty clearly is why Europe is such a mess (besides inflexible labor markets) – the insane idea of allowing some of those countries to borrow and spend on the strength of the D-Mark (one can only hope that was more about politics than economics). There is no easy way out of the aftermath of that.

    Speaking of Nobel Prizes, interesting to contrast Tirole’s work with that of another recipient, Buchanan.

  60. 60 60 Daniel

    “Like today we know when a big stimulus doesn’t work that proves it should have been much bigger.”

    You mean when we have a predicted insufficient stimulus (by Krugman Romer, and several other prominent economist)

    “I think the consensus still believed in an exploitable Phillips Curve long past the 20s, which as you note required an intentional recession to fix.”

    Believing in a Philips curve wasn’t the problem in the 20’s. Overestimating the threat of inflation and overreacting was.

    “At least one thing we should know pretty clearly is why Europe is such a mess (besides inflexible labor markets) – the insane idea of allowing some of those countries to borrow and spend on the strength of the D-Mark (one can only hope that was more about politics than economics).”

    We’re in partial agreement here, but the focus should be on the national fiscal policy with international monetary policy. It isn’t a matter of saints in Germany and devils in the periphery, it’s a matter of no automatic stabilizers from a central fiscal government that can control it’s currency. That’s why we don’t have a problem between the imbalance between the strong North Eastern dollar and the weak Southeastern dollar in the United States. The differences are similar to the differences between the core and periphery countries in Europe but we have a centralized fiscal union to stabilize the hole.

  61. 61 61 GabbyD

    Did doctors promise, implicitly, that they could completely prevent breakouts of disease?

    i dont think so.

    at best, doctors and epidemiology promises to understand how breakouts happen, and how to stem them once they occur. thats all.

  62. 62 62 iceman

    60 – and of course there are other prominent economists…all we know objectively is that ‘partial’ measures did not produce partially positive results as advertised, leaving proponents to defend a negative i.e. that things would’ve been even worse. The only thing that seems clear is that it’s too soon to be rendering verdict or declaring consensus.

    – stagflation was real, and the fact that even a clearly discernible statistical relationship can break down when we try to act upon it should give pause to any fine-tuner.

    – agreed but if a central European govt wasn’t in the cards (and probably never will be), Maastricht criteria are going to be ineffectual, and yet the union was expanded to more marginal members anyway, then perhaps the whole experiment was doomed from the start and in a foreseeable way.

  63. 63 63 Daniel

    “leaving proponents to defend a negative i.e. that things would’ve been even worse.”

    No, we can use the counterfactual between us and Europe now, us and Japan in the lost decade, and us and us during the great depression. By all measures we’re doing much better because we’ve figured out some more about how to use monetary policy effectively.

    “stagflation was real, and the fact that even a clearly discernible statistical relationship can break down when we try to act upon it should give pause to any fine-tuner.”

    Yes stagflation did exist, but that doesn’t mean you know why it existed. You use one event to then say that we can predict the reasons for stagflation, and then say that monetary policy can have no real effects?

  64. 64 64 Daniel

    @ iceman,

    And there is some pretty good evidence that the stimulus stabilized private sector employment growth. Mainly that private sector returned to normal growth around the time that the stimulus was in full throttle mode:
    http://data.bls.gov/timeseries/CES0500000001

    But then, after the the middle of 2010 public employment took a huge nose dive (so there was actually very little and even negative stimulus at this point).

    http://data.bls.gov/timeseries/CES9000000001

    So while the private sector was stabilized, the public sector loss in employment was great enough to keep us essentially at a flat line. If government employment was somehow crowding out investment, why didn’t we see a spike in the slope of private sector employment around the time when the stimulus starting turning negative?

  65. 65 65 iceman

    I didn’t say monetary policy can have no real effects (perhaps people can be fooled in a predictable way). But does it concern you at all that even if we really “know” how to smooth out a downturn, when you step back it seems like the LR does indeed become a series of SRs, with each crisis and response setting the stage for the next, bigger round? E.g. Latam banks to 87 crash to peso to LTCM to dot-com to housing bubble to…some call this moral hazard / Fed put. Setting aside LR-impact of debt overhang of course

    We have I think had sufficient time to analyze stagflation and conclude that failing to properly account for expectations was a major gap. Some would think even as we continue to learn (from our mistakes?), this kind of example calls for doing so with a proper measure of humility e.g. Hippocratic oath.

    And I would certainly hope there is some evidence that deficit spending $800B did *something*, which is not to say it passes a cost-benefit…unless of course you truly believe in this case the borrowing pays for itself, as some have argued (unfortunately tend to be same people who ridicule similarly extreme supply-side views)

  66. 66 66 Daniel

    I’d say that when we didn’t have a federal reserve system, we still lurched from crisis to crisis. I think business cycles are just a consequence of having a banking system, and the federal reserve can help mitigate the variance of the shocks and lower overall pain. You can see this from the very long crisis in the 1800’s when a federal reserve system did not exist.

    http://www.nber.org/cycles.html

    I agree, incorporating expectations into monetary policy was an important development, but this development did not lead us to stop using monetary policy all together. The Taylor rule is a monetarist rule. And yes Alan Greenspan had a hand in creating the housing bubble, but what do you expect from a Laissez Faire who barely believes in monetary policy himself and is basically a Republican frontman (which you could see by his utter obsession with debt during the Clinton administration, and his turn the other cheek policy during the Bush years). This doesn’t mean that we shouldn’t use monetary policy, it means we should elect better federal reserve chairpersons.

    Okay, so I produced evidence showing an impact, and now you say well “I certainly hope there’s some evidence for spending $800 B”. I think you’re dodging now. I think you’re falling into the trap of being fooled by an inconceivably large number by human standards, when in reality it’s just 5% of one years income (not a large amount of additional debt even if you consider the US government to behave like a responsible household, which you should not, since governments have much more leeway in debt since they have no expiration date and can produce their own currency). Keynesians also have a long track record for talking about the dangers of longrun sustained unemployment effects on labor supply, so I don’t know what you’re talking about. Keynesians don’t think that supply-side recessions don’t occur, just that certain recessions are mainly demand-side, while others are mainly supply-side and the current one we are now in isly main demand-side. Given that you don’t know this distinction I seriously doubt you’ve spent much time reading Keynesian arguments because you’ve a prior decided that it’s trash.

  67. 67 67 iceman

    Daniel – sorry for the delay. I enjoy learning from these chats, notwithstanding your penchant for making extreme caricaturizations (I believe you call them straw men), like my suggesting recent policy actions produced disappointing results implies that monetary policy is useless or Keynesian arguments are trash. (I may have phrased a few things a bit provocatively myself for effect, but with respect to your charts, I would say they seem to show a negative correlation, and in any event do not depict generally strong or even “normal” growth?) Or concluding people don’t understand basic things like the difference between S&D. If I’m unclear on a point perhaps you could ask me to clarify. On another thread people are saying John Taylor lacks a basic understanding of monetary economics – ridiculous.

    As you know the topic here and the over-arching debate at least since the General Theory has been when (and how) to intervene in light of potential longer-run consequences of ST smoothing e.g. borrowing against future growth. Here’s what I currently think:

    – These are not easy questions for which there are obvious answers, and as we continue to learn (which is good) our policy prescriptions should recognize the uncertainties and limits of our current state of knowledge (in a difficult field which does not lend itself to controlled experimentation) — as well as a natural *political* incentive to over-weight the SR.
    – Stagflation in particular taught us important lessons about the limits of fine-tuning (which had by that time gone well beyond what Keynes had prescribed).
    – I see perhaps two instances of true liquidity traps in the past 100 years (do you see more?).
    – I see several instances of bailouts / policy (primarily liquidity-based) responses in the 80s and 90s alone (I left out the S&L bailout), which cut across both major parties and gave rise to the term “Fed put”, which in my view contributed importantly to the ‘great recession’.
    – In my view hysteresis is a relatively recent addition to the new Keynesian arsenal, and an important one as it attempts to speak directly to the longer-run issues. Personally at this point I can’t help feeling a bit skeptical that it is overstated as a convenient pretext for more activist policy, e.g. in the context of arguments that the borrowing pays for itself (just as many were justifiably skeptical of *analogous* (but distinct) supply-side claims).

    I would welcome your thoughtful responses to any of these statements.

  68. 68 68 Daniel

    @Iceman,

    I enjoy conversing as well. I agree that we shouldn’t always intervene, just that if there’s ever a time to intervene, this time would be it. Can you agree with that at least? I also believe the fed does more good than harm for the reasons mentioned earlier but that’s definitely not a settled matter. I do however feel there’s more positive evidence on monetary than fiscal policy since the types of recessions the fed can mitigate the variance for are more frequent than liquidity trap type recessions. I tend to think that you’re putting too much weight on Fed put and not enough on lack of bank regulation for the great recession, but both could have contributed.

    Where exactly did you see negative correlation in my graphs again? I showed you a stabilization in private sector employment at the peak of stimulus (that’s exactly what stimulus is supposed to do, offset the irrationally large pull back in the IS curve).

  69. 69 69 iceman

    It always feels like we have to do something, which is why I wished we could somehow credibly commit to saying “this is the last bailout ever”.
    It’s also fair at this point to wonder if we would’ve had a quicker recovery by doing less. Some point to the early 20s as an example.
    I agree IF we are going to subsidize banks we require some ability to regulate.
    It’s also possible the stimulus would’ve worked better had it not gone largely to shoring up municipal balance sheets (seems similar to shoring up banks’ b/s). Some non-Keynesians said let’s at least do something with a decent multiplier like a payroll tax holiday. “Shovel ready” was not so much. But of course sub-optimal implementation probably needs to be factored into any spending stimulus.
    Your charts – you’re looking at the blip I’m looking at the general picture.

  70. 70 70 Daniel

    “Some non-Keynesians said let’s at least do something with a decent multiplier like a payroll tax holiday. “Shovel ready” was not so much.”

    We did, that was the cut in the social security tax.

    ““Shovel ready” was not so much.”

    I agree, a better use of funds for infrastructure would have been to give a lump sum to each state proportional to its population to ease the budget constraints on States. This would have been 100x easier to implement and no one could have questioned the allocation of funds. This is what happened during the Reagan recessions, so I wasn’t sure why we needed to create new programs other than to make it look like “new” things were happening.

    “Your charts – you’re looking at the blip I’m looking at the general picture.”

    I’m not sure what blip you think I was talking about, it’s clear that private sector employment returned to it’s normal rate of growth path when the stimulus had taken full affect, while public sector employment started declining afterwards (when the financial sector was no longer fearing an implosion of their system). This is what leads to the sideways growth in jobs. My question to you was explicit and you’ve never really answered it. If the government doing less would have made us better off, why didn’t the decline in public sector employment lead to an increase in the slope of private sector employment?

  71. 71 71 Daniel
  72. 72 72 iceman

    Article primarily credits Fed policy – wouldn’t Krugman et al disagree? And again at this point it seems quite fair to wonder whether we would’ve had a similar (or quicker) recovery without the deficit spending (and possibly without QE, at least the follow-on rounds, a point with which I presume Krugman et al would agree). I’d be interested in your take on why, say, the episode in the early 1920s where we had a rapid recovery with little policy activism is not a good comparison.

    I confess I’m struggling a bit to see the point you’re making with your graphs and how to respond to it, in part because I didn’t think I was making an argument based specifically on crowding out, I just think at this point to any impartial observer it would seem the stimulus was less effective than was hoped e.g. based on the UR forecasts. Maybe too much of it was poorly spent, maybe expectations are affected b/c people see the debt overhang +/or know we’re just front-loading activity…
    But I will say in the graphs I see private start falling as public continued to rise, then private bottomed out and began to rebound *before* the spike in public began – so is the theory that the private sector *anticipated* the policy effects in a positive way? From there private continued to rise thru both the public spike and extended decline (so the majority of the graph does show a negative correlation). It seems your thesis rests on characterizing this as ‘merely’ a resumption of “normal” private growth — with “normal” being the pre-2008 housing bubble years? Of course under the circumstances perhaps re-establishing that slope was pretty darn good? And couldn’t a more modest / normal trend be consistent with the notion that the stimulus is partly offset by longer-term expectational factors?

  73. 73 73 iceman

    More succintly – I’m suggesting private did what it was going to do anyway (but public spent $800b in the process). Your thesis seems to be that the stimulus must have worked *because* private did *not* grow faster. Have a good weekend

  74. 74 74 Daniel

    @Iceman,

    No Krugman would not agree that we would have had a faster recovery if followup QE did not occur, his argument was that it had a small to null effect because that’s what happens with monetary policy when you have a flat LM curve. That said, it had no mechanism to hurt us with a flat LM curve.

    So what you’re seeing when public sector growth is growing prior to the drop in public sector is normal public sector growth related to population. The collapse in private sector jobs was due to a major shock in the banking sector something you’d be hard pressed to find a link with public sector growth for.

    For why the private sector stabilized prior to that bug surge in public sector one possibility us that remember the stimulus was half tax cuts which occurred prior to the surge in public sector employment. So this likely had stabilization effects as well. You could make the argument that the tax cuts were a more effective stimulus. My argument is that if the private sector was going to do what it did regardless of public sector employment we could have had a faster recovery simply by keeping public sector employment flat or even faster if we had kept it growing on a normal growth path.

    For the 1920’s, I think I answered this already. Not all recessions are the same. Milton Friedman believed that a large chunk of this recession could be due to Fed reacting too strongly to inflation. Also, we had a large surge in labor force participation. This is completely different than a financial collapse. We were notbat the ZLB. When the Fed let up the recession ended. A better comparison would be the 1980’s (also a Fed caused recession but much less deep because they correctly calculated the change in inflation expectations they needed.

  1. 1 A Little Perspective at Steven Landsburg | The Big Questions: Tackling the Problems of Philosophy with Ideas from Mathematics, Economics, and Physics | Daniel J. Smith
  2. 2 A Little Perspective | Brucetheeconomist's Blog

Leave a Reply