Economics should think small


I attended a fascinating, if somewhat disturbing talk at the Cranlana Programme by Scott Borg, director and chief economist at the US Cyber Consequences Unit, who is a cyber warfare specialist and economist at the US Cyber Consequences Unit.

It included horrible descriptions of the end of the world as we know it, much as expected. It turns out that mutually assured destruction exists just as much in the cyber realm as in the nuclear realm.

But Borg was also exploring an economic approach that looks like it starts from much sounder premises that most economic ideas. As he commented to me, most macro economic theory is borrowed from eighteenth century physics. He said “One of the nice things about the value creation stuff that I and my colleagues do is that we started from scratch. We are not importing a bunch of concepts and mathematics from physics.”

Here are excerpts from my interview with him, in which I think he showed an unusual degree of common sense. Borg very much concentrates on micro economics. He is reluctant even to comment on macro economics:

“Micro economics can identify mechanisms by which things operate. And so it is certainly a science. A lot of macro economics is pretty borderline because it is not identifying mechanisms it is looking for statistical correlations. It is a rule of thumb that if you have statistical correlations only you are not really dealing with a science.

A science doesn’t need statistical correlations because it can say absolutely what happens and how it happens. If I am doing chemistry I don’t have to say there is a statistical correlation that suggests this reaction. We have got this reaction or we don’t. Whenever people are talking statistics they are usually not there as a science.”

Nice to hear someone say it. I asked him about his view of comparative advantage, which Paul Krugman has described as a “badge of honour” for economists (he mainly means macro economists). It has been the mantra behind free trade for decades. Borg argues essentially that it has legitimacy at the micro level, although it needs to be refined. But that’s about it. He is seeking to answer the question that really maters: how to economies grow? And unlike those would be eighteenth century physics, he does not assume a steady state system:

“Countries should seize opportunities to create value. They should not fall into some stereotyped idea that someone has foisted on them of what they are good at and stick to that. That would preclude all kind of economic ventures. So basically comparative advantage as a concept is incontestable. Comparative advantage (as a dictum that countries should only specialise in what they are good at) is a great fallacy.

Countries and companies and individuals should seize opportunities to create value. The value that you can create depends on what you can do that your competitor can’t. If you can’t do something better than your competitor you are not creating any value at all. It is better to have you out of the game.

There is a better way to describe these economic points. It is to think in terms of value creation. The value that is created by an activity is the opportunity, it is the willingness to pay of the customers minus the opportunity cost to suppliers. The willingness to pay is based on the best alternative is for those customers if they don’t do that particular deal. The opportunity cost is based on what the best opportunity is for the suppliers if they don’t do that particular deal. So to create value you need to have a gap between the willingness to pay the opportunity cost and that amount of value creating is always done on a comparative basis. It is always based on what the other alternatives are.

When people use phrases like comparative advantage they are trying to look at one picture, one piece of that gestalt of that bigger picture and not look at the whole picture.

(Nobel prize winner) Paul Romer doesn’t really have a theory of economic growth or innovation. What he did was he fiddled with the equations so that there is another variable in there it is a little bit rearranged. But the model is just fiddling with the equation, it is not an actual theory of innovation or economic growth.

Adam Brandenburger from NYU and Harbone Stewart (are developing these ideas using) co-operative game theory based value creation. A lot of what people have heard about complementary value chains and value this and value that comes out of work that was pioneered by Brandenburger and Stewart.”

Borg says the cluster theory developed by Michael Porter, another attempt to explain growth at a micro economic level is “all very fine it is not very rigorous:

” It is fine as far as it goes. We need a lot finer grained understanding of that now and the reason I cite B and S breakthrough articles that you can use to build a better understanding. Cyber security it has enabled me to do a lot of new economic theory. Cyber attacks can mess with anything in such a fine grained way that you can’t do any of this wave the hand stuff any more. You can’t approximate a willingness to pay curve, you can’t call it a demand curve and just draw it in some kind of vague smooth line. You have to appreciate that it is a step curve and each step is determined by what the alternatives are to using that product. The same way with an opportunity cost curve, you have to start looking at that much more analytically.

The mechanisms within that market by which people negotiate their way in also can’t be just done with a wave of the hand, you can’t just assume they will actually do that. You have to start looking how that market functions in the real world. That allows much finer grained micro economics and it make a lot of what people are taught as micro economics in undergraduate text books so crude as to be misleading. Even wrong.”

Latest posts by __ADAM__ (see all)


  1. There is nothing new in these points. Microfounded macro models have been around since the 70s. Google the Lucas critique. Most macro is microfounded these days specifically because of the problems inherent in drawing conclusions from observed correlations in aggregate historical data.

    Honest question here, how many macroeconomic papers have you actually read? You seem delighted by the fact that an economist is ‘finally’ pointing out something that economists have been aware of and worked on for decades. It implies a fair degree of ignorance I’m afraid to say. There is nothing wrong with ignorance, so long as it is coupled with humility. Constantly railing against the shortcomings of economists when you are poorly informed about what they actually do is not a mark of humility.

    • Quite a few, but a better question would be to ask me how many macro economists have I interviewed over 30 odd years? A very great many, including many overseas economists. Also worked with a few in the financial markets and saw how reliable their predictions are (you know, the bit that scientists are supposed to be able to do). My particular favourite was watching an economist for a major broker give his predictions to the institutional brokers then go back to the fixed interest department and do the exact opposite, aware that what was supposed to happen in the markets, according to general equilibrium or whatever, was exactly what everyone had factored in and was trying to outplay. He did quite well. The institutional brokers, meanwhile, used to ignore him completely.

    • I might add, for what it is worth, that I am not ignorant of epistemology, having studied it as part of my PhD, and the point I am making is epistemological. The result may be repetitive, but that is because I think its significance is underestimated. If economists do admit that what they are doing is not a science (haven’t seen many), then what is it? That is an epistemological problem, i.e. what kind of knowledge is possible here? I think there are possible answers, but I do not see the question being posed. Instead, what I see is a complete lack of humility from economists — understandably enough, I suppose, given that they could hardly get paid a salary or fee for saying they do not know. So instead they become a bit like modern day astrologers using transactional language instead of planetary movements. It is not just an academic point. That transactional language rules the world, shapes so much of our behaviour and value system. I guess it is not something that should be emphasised on an economics blog …

    • “There is nothing wrong with ignorance, so long as it is coupled with humility.”

      Learning is a grace, particularly when coupled with modesty; whereas a little knowledge coupled with arrogance merely empties the lecture hall.

  2. So, wait a minute, you just endorsed an economist who described microeconomics as a science but now you are telling me that any economist who describes economics as a science is as bad as an astrologer? Such a characterisation, by the way, is almost too baseless and narrow-minded to warrant a rebuttal. There is no basis to astrology whatsoever. There is, however, very definite relationships between economic policy and economic outcomes. Studying and understanding these relationships yields results that have a material impact on human welfare, as evidenced by say, the standard of living in North Korea and South Korea, or the elevation of hundreds of millions of Chinese out of abject poverty in a generation. Every day contributors to this blog toil in an effort to understand, predict and provide advice on the workings of our economy. How you could possibly assert that they suffer a collective delusion and offer insights no different to an average astrologer is beyond me.

    Further compounding my confusion is the fact that you have made it your personal crusade to expose the sham that is the use of mathematics in economics (without ever providing specifics), but here you are telling us that economists need to focus on microeconomics and ignore the folly of statistical inference at the macro level. The bizarre part of this argument is that, historically, microeconomics was where you would find the maths; so far as I am aware the earliest use of calculus was in Johann Heinrich von Thunen’s work in the mid 19th century. Macro didn’t really exist until the invention of national accounting in the 20th century. It was (by today’s standards) relatively devoid of maths and followed more of a newspaper article-style of analysis that you might today find in financial media. Post-war, it also relied heavily on statistical inference to form policy decisions. The microfoundations/rational expectations revolution was where macro moved decisively away from the ‘living’ written word and towards the ‘dead’ mathematical symbol, and lessened its faith in statistics. So I really cannot wrap my head around what your position is exactly. You don’t like maths, but you happily advocate micro and game theory, which are basically mathematical descriptions of human behaviour, and you don’t like statistics but prefer non-mathematical, written-word macroeconomics, which by definition requires statistics if it is to say anything at all.

  3. Back before “economics” and “political economy”, some philosophers studied “dialectics”, which was very like how Borg describes his insights: “Micro economics can identify mechanisms by which things operate…..”

    Adam Smith’s “The Wealth of Nations” is really a work of dialectics. So are many other classical economists works. They explain how things work in the real world.

    So I think “economics” really lost track of its origins, in becoming all formulaic and “macro”.

    • +100. You are absolutely right. Unfortunately today many people are discovering the wheel again and again when the reality gets ugly right before their eyes. So Borg isn’t an isolated case at all. He is more of a common rule.

      Back to the future, this is the only way to learn….

  4. Free_Market_Delusion

    Well all of that article made perfect sense to me.

    I did an economics degree and a science degree simultaneously.

    Going from chem lecture to an economics lecture showed to me absolutely abundantly clearly that economics is not a science.

    • But Borg seems to me to be saying, that being able to describe what will happen at the “micro” level in market interactions, really is equivalent to “science”. I believe this. A lot of what Adam Smith did was nothing more than this. His writings read like a whole lot of astute observations about human nature. In fact he would not have even been thinking he was part of a profession like today’s economists do. You can read Edmund Burke, followed by Smith’s “Theory of Moral Sentiments”, followed by “The Wealth of Nations”, with a strong sense of connectedness.

      Reading Alan W. Evans’ works on Land Economics was a breath of fresh air, because he gets right down to the level of the individual land owner, the developer, the local government, and the investor/speculator. He then develops explanatory theories from there.

      But mainstream economics, right up to today, after years of global crisis stemming from bubbles in urban land prices, remains utterly incurious about the “micro” level of urban land markets, and remains utterly clueless when it comes to recommending policy responses.

      Their approach is via macroeconomic equilibrium models, and they explain a bubble in house prices as being “due to the fundamentals of supply and demand”, then conclude that there is nothing to worry about…..!

      They are clueless about the consequences of distortions in demand, and they are clueless about the effect of debt expansion. Saying that house prices at a median multiple of 9 to 11 are nothing to worry about “because they are explained by fundamentals of supply and demand”, when supply is grossly distorted, is like saying “don’t worry, our troops are only losing the war because they are out of ammunition, their tactics and bravery are fine”, and not bothering to do anything about the supply of ammunition…..!

      • “….our troops are only losing the war because they are out of ammunition, their tactics and bravery are fine”, and not bothering to do anything about the supply of ammunition…..!”

        Very good.

  5. JohnLounsbury

    Good discussion.

    The quote from Borg, however, is completely too shallow on his part:

    “It is a rule of thumb that if you have statistical correlations only you are not really dealing with a science.

    A science doesn’t need statistical correlations because it can say absolutely what happens and how it happens. If I am doing chemistry I don’t have to say there is a statistical correlation that suggests this reaction. We have got this reaction or we don’t. Whenever people are talking statistics they are usually not there as a science.”

    He has unfortunately followed physics only through the 19th century. The entire development of particle physics in the 20th century is a study of probability distributions and statistics in order to reconcile the observed “macro” physical world with the mathematical descriptions of the “micro” physical world of particulates.

    In a philosophical sense, if not yet in a operational sense, this is a development gap that Lucas and others have sought to bridge. Unfortunately the mathematical tools applied, the analytical thinking used and the inappropriate assumptions made have resulted in dismal failure. Note: Many macroeconomists will want to debate my assessment.

    But I’ll stay with it.

    • That is right and what matters in economics and other social areas is the method of analysis. If the method isn’t adequate to the always changing socioeconomic reality then the results are no more than wishful statements, no matter how much math and stats one have used to support arguments totally detached from the real root of the problem and even more detached from its healthy solution.

  6. surfbeach2536

    comparative advantage expanded to account for opportunity cost is of course only the starting point.

    If we then consider the “supplier” is the community (country) as a whole then the opportunity cost should include the cost of underemployment on top of all of the micro mechanisms.

    We nned to remember the science (micro) is only knowledge, application of that knowledge will show how wise we are

  7. The economy is not science. Economists failed. The GFC caught them all, except a few, with their pants down. Economists talk emotions, the emotions of supply and demand, the emotions of fear and greed.
    If we regard them as scientists, then we should include them with meteorologists, who can be wrong 80% of the time and still hold a job.

    • Yeah, but……

      I think the problem is that economists who DO make accurate predictions, are not popular…..!

      How many of these guys have been appointed to top administrative jobs, eg in the US government or Federal Reserve Bank, after the crash, seeing they have been proved right?

      None of them. All the top jobs are still filled by people who GOT IT WRONG. And the chances are their policy recommendations to “solve” the problem are STILL wrong.

      Possibly we need to give the public the chance to actually vote for someone like Peter Schiff versus Ben Bernancke as Chairman of the Federal Reserve……?

    • Actually meterologists have been a relative succsss, weather predictions have increased in accuracy %300 in the last 25 years …

      … could an economist say the same ?

  8. Ι’m not sure if economics is a failed science, or it never was a science in the first place, or it is irrelevant because the game is rigged anyway, but to me the end result is the same: economics is useful for very simple stuff only, like GDP deflators and such. But to understand what’s really going on and why things move the way they do I find more useful and reliable answers in other sciences, mostly anthropology, philosophy and history. I feel I understood more about money and debt from David Greaeber and his debunking of Adam Smith than I did studying economics at Uni. Same for Karl Polanyi, Wallerstein, Hobsbawm etc. I’m against using “advanced” economics for making important decisions, it’s just too dangerous