What Austrian business cycle theory does and does not claim as true

Last month the FT’s Martin Wolf asked a simple question, “Does Austrian economics understand financial crises better than other schools of thought?” After admitting that neo-classical models did “a poor job in predicting the crisis and in suggesting what should be done in response”, he points out that the following Austrian arguments have held up well: “inflation-targeting is inherently destabilising; that fractional reserve banking creates unmanageable credit booms; and that the resulting global ‘malinvestment’ explains the subsequent financial crash.”

Unfortunately Wolf goes on to make the error of confusing the causes of the crisis to the policy debate regarding the recovery: “Austrians also say – as their predecessors said in the 1930s – that the right response is to let everything rotten be liquidated, while continuing to balance the budget as the economy implodes. I find this unconvincing.”


Firstly, if you are looking towards Austrian business cycle theory to provide a complete theoretical explanation for (i) the artificial boom, (ii) the economic recession and (iii) the appropriate policy response to generate new growth, you may well be disappointed. But if it is unreasonable to expect one (relatively unknown) school of thought to unambiguously settle each of these issues, it is also unreasonable to reject the parts that do stand up to scrutiny purely because they don’t explain everything.

The Austrian insights are predominantly a theory of unsustainable credit-induced booms. Therefore they are not equally applicable to all “boom-bust” cycles, and originally didn’t even attempt to investigate the recovery process (Hayek labelled this part the “secondary deflation”, implying that something else – the malinvestment of capital goods – was the primary problem). The famous “Austrian” histories of the Great Depression were more concerned with the boom than the bust – Lionel Robbins’ The Great Depression was published in 1934, and Murray Rothbard’s America’s Great Depression stopped at 1932. I would suggest that Wolf stops viewing Austrian ideas as substitutes for all other explanations, and takes a more opportunistic view – Austrian ideas explain the boom, Keynesian and monetarist ideas are also required to explain the bust.

Secondly, Wolf misrepresents what Austrians do say about depressions. The “liquidate” thesis is false, since many Austrians acknowledge that avoiding a monetary contraction would prevent a deflationary spiral leading to depression. Because of the complexity of this policy, the argument tends to shift from monetary policy to monetary regimes, as Austrians do have a positive programme for banking reforms that would prevent future crises. The problem is that only Austrians confront head on the harsh economic reality that there is no such thing as a free lunch. Once malinvestments are made, they are costly to correct. This does not imply that politicians “do nothing” necessarily; just that efforts are concentrated around the ability of markets to function as they should.

Indeed, consider Paul Krugman’s response to Wolf’s question, “why isn’t there similar unemployment during the boom, as workers are transferred into investment goods production?” Well there is unemployment during the boom – according to a recent study, 2.65m private sector jobs are lost every year in the UK (but we don’t tend to notice because even more are being created). Focusing on the “aggregate” level of employment misses the relative adjustments that are being made, as people move between jobs due to changing economic conditions. Indeed Krugman also seems to be implying that the “boom” should correspond with “overinvestment”. Whilst this isn’t entirely incorrect, the Austrian argument is that it leads to “malinvesment” – subtleties that aggregate variables gloss over.

The problem is that neoclassical economists believe in a circular flow model that abstracts away from time. Austrians, by contrast, appreciate that capital investment occurs over time. As Roger Garrison says, “the specificity and durability of the long-term capital does not allow for a general timely reversal.” Or as Arnold Kling says, “booms are slow and crashes are sudden” (note that John Hicks made the same error in his criticism of the Austrians; it seems that Krugman is unaware of this and the secondary literature it has generated).

In short, whether Austrian ideas have something to add depends on whether you view the pre-crisis economy as fundamentally sound. As Garrison points out, there are two alternative views: “did the collapse occur (a) in the midst of a period of healthy growth because of sheer ineptness of the central bank or (b) near the end of a policy-induced boom that was unsustainable in any event and in the midst of confusion about just what the problem was and how best to deal with it?”

If you answer (a) you’re a monetarist, and there’s no surprise that Austrian ideas seem alien. But if you believe (b) then I would encourage you to learn more about the economic theory that explains economic crises so majestically.

39 thoughts on “What Austrian business cycle theory does and does not claim as true”

  1. Posted 25/05/2010 at 12:36 | Permalink

    […] IEA Blog » What Austrian business cycle theory does and does not claim as true By Dr. Anthony J. Evans, on 25 May 10 IEA Blog » Blog Archive » What Austrian business cycle theory does and does not claim as true. […]

  2. Posted 25/05/2010 at 16:08 | Permalink

    Austrianism is so stone-age when it comes to business cycle theory.

    In the 1950s Jay Forrester of MIT invented an experimental game called the Beer Distribution Game, which demonstrates how human behaviour and dynamic structure can produce endogenous oscillations, generating a cycle as an emergent property of a complex system.

    Problem number 2 with Austrian theory as set out in your Primer on von Mises is that it assumes that, if you push the interest rate below originary interest (that which reflects time preference), then businesses are liable to overborrow. This depends on the unwarranted assumption that all boom-finance is based on debt and none on equity.


  3. Posted 25/05/2010 at 16:12 | Permalink


    It also assumes that ramming the interest rate down to the floor is by itself sufficient to stimulate investment, malinvestment or over-investment however you want to call it. It could do so only if it fell below the marginal efficiency of capital, and that variable depends on the present value of the expected income stream from the investment and the supply price of the asset.

  4. Posted 25/05/2010 at 21:16 | Permalink

    Michael – I am afraid I do not understand problems 1 and 3 as stated. However, with 2, what you say simply is not the case. The money creation that results from holding interest rates down can be invested in equity assets (encouraging the issue of new equity finance) and the expanation of equity finance will also be encouraged in the boom industries. It is true that both these mechanisms deny what neo-classicals would call “rational expectations”, but then much of Austrian theory is based on the limitations of knowledge and how entrepreneurs use local and relevant information – loose monetary policy distorts that information.

  5. Posted 25/05/2010 at 21:30 | Permalink

    […] as true By Steven Baker MP, on 25 May 10 Founding Fellow Dr Anthony J Evans blogs for the IEA: What Austrian business cycle theory does and does not claim as true: Last month the FT’s Martin Wolf asked a simple question, “Does Austrian economics understand […]

  6. Posted 26/05/2010 at 06:12 | Permalink

    Problem No. 1 – the Beer Distribution Game – is a simulation of supply chain management. Where there is a supply chain with inventory kept at each stage, and a time delay between the placing of an order and its fulfilment, then a small one-off change in consumer demand will trigger oscillations throughout the supply chain. These never settle down to a simple flatline.

    Problem No. 3 – The marginal efficiency of capitalis the rate of discount which makes the present value of an expected income stream just equal to the supply price of the asset that generates it.

    If, and only if, the present value is greater than the asset price, then the purchase of the asset is worthwhile.


  7. Posted 26/05/2010 at 06:22 | Permalink


    If expectations of future revenue are low, then even a tiny interest rate may not be sufficient to generate a present value sufficiently high to clear the cost of an income-yielding asset.

    So we have an answer to Problem 2. Money is created by fractional reserve banks as they lend, not in the course of financing by equity. Cutting interest rates might induce them to increase the quantity of loan capital supplied, as they need to lend more in order to earn the same interest income. But that doesn’t entail that there will be the demand for it. That is critically dependent on expectations.

  8. Posted 26/05/2010 at 09:51 | Permalink

    Michael – The Beer game is an excellent pedagogical device but in the spirit of my article I would suggest that the possibility of endogenous cyclical phenomena do not disprove the possibility of credit-induced cycles. Indeed efforts to study the inventory management of organizations brings in the elements of time and capital that is so crucial to the Austrian cycle. The two could work together.
    You also seem to be attributing too much to interest rates as opposed to the money supply itself. I have to ask – are you familiar with the likes of Roger Garrison, Steve Horwitz and Tyler Cowen?

  9. Posted 26/05/2010 at 10:08 | Permalink

    Surprised also at Michael’s use of problem 1. Both classical and Austrian theory very much emphasises the possibility of disruptions, cycles, an allocation of resources that is not optimal ex post because of entrepreneurial error and fundamental uncertainty. They would argue that this is a reason to minimise government sources of such errors and disruptions not to dismiss them as a possibility.

  10. Posted 26/05/2010 at 10:16 | Permalink

    Reply to posts 8 and 9. The advantage of the Beer Game approach is that it enables the isolation of cycles in the flow of circulating capital, leaving aside the role played by credit and fixed capital investment.

    I can’t say I’ve heard of Garrison, Horwitz and Cowen.

  11. Posted 26/05/2010 at 10:26 | Permalink

    Michael – granted, cycles may occur without excess credit creation, but that doesn’t mean that excess credit creation won’t amplify such naturally occurring cyclical behaviour. Again, these aren’t either/or explanations.

    Regarding Garrison, Horwitz and Cowen – that makes sense. There have been a number of books written on Austrian cycle theory (some in support, some more critical) that came out around 2000. It’s no wonder you dismiss it as “stone age” if you’re unaware of the recent work in the field.

    I would encourage you to familiarise yourself with the contemporary literature before being so dismissive. The debate has moved on since the 1950s!

  12. Posted 26/05/2010 at 11:50 | Permalink

    The problem with the theory of credit-generated cycles is this. In a fractional-reserve banking system, money is created as it is lent, and is destroyed as the loan is paid off. Many businesses which sold into the boom can do this and remain solvent, in which case net money creation over the loan repayment period within the upswing of the cycle is zero. Secondly, the concept of ‘malinvestment’ is itself problematic because of the fact of depreciation and of a second-hand market in fixed capital. If I invest well in three machines and malinvest in a fourth, then the malinvestment self-corrects with the passage of time as soon as one of the machines wears out.


  13. Posted 26/05/2010 at 11:56 | Permalink


    Alternatively, if I find that I have overinvested in a machine, I can sell it to another business for which it would not represent a malinvestment at the second hand price. On the whole, it seems that malinvestment doesn’t matter much as long as businesses remain solvent.

    Finally, credit is not likely to be created unless a bargain is struck between the borrower and the lender. That requires concurrence that the business plan is sound for debt repayment on the basis of expectations of revenue ex ante.

  14. Posted 26/05/2010 at 13:27 | Permalink

    Michael – capital goods are heterogeneous. We can’t simply substitute them between business plans. There is a real cost involved in resource reallignment. When you say “the malinvestment self-corrects with the passage of time” you are right, but this “passage of time” is what we otherwise call a “recession”.

  15. Posted 26/05/2010 at 14:43 | Permalink

    True, capital goods are heterogenous and hence mutually incommensurable. That’s central to Piero Sraffa’s capital theory. When I say that modest malinvestment corrects itself with the passage of time I am referring to depreciation of fixed capital. If I invest in four machines when I only need three, then I can mothball one until another wears out.

    That’s a case of premature investment, rather than pure malinvestment. It obviates the need to buy the fourth machine later.

    This has implications if the upswing of the boom is longer than the useful life of the machine. It’s not the subject matter of a recession.

  16. Posted 26/05/2010 at 22:26 | Permalink

    If you define “modest” malinvestment as merely “premature” investment then I suggest you read up on how capital theory relates to the trade cycle, and understand how – in reality – they differ.

  17. Posted 27/05/2010 at 10:40 | Permalink

    I question the notion of ‘malinvestment’ from first principles. Is there in the real world a unique solution in terms of a Pareto efficient state of the economy in relation to which an investment could be ‘mal’?

    If not, is there a range of ‘right answers’ to the economic problem within which an investment solution could fall and so be economically acceptable?

    The line I’m thinking along is this. An economy is a complex system. The market, as a process, drives out redundancy as it tends to allocational efficiency. This reduces resilience and multiplies the number of critical failure points within the system. If one of them fails, then the system collapses.

  18. Posted 27/05/2010 at 12:30 | Permalink

    “This depends on the unwarranted assumption that all boom-finance is based on debt and none on equity.”
    I dont think this is right. Austrians such as Frank Shostak fully accept that equity will play a role in expanding investment. But the low interest rates will make unsustainable investments attractive (cos of the fall in the cost of capital). Equity, along with increased debt, will flow into and be wasted on such unsustainable investments. Once the bust comes society will be poorer since that pool of funding (or real wealth if you like) will have been frittered away, all to save the face of political leaders and central bankers.
    Thanks Gordon!

  19. Posted 27/05/2010 at 12:41 | Permalink

    “I question the notion of ‘malinvestment’ from first principles”.

    Investment needs to be balanced by savings. Savings form the pool of funding for the investments carried out by entrepreneurs. This ensures that there is a balance between spend on consumer products and the demand for the output of the capital goods sector (CGS). The boom phase expansion of the CGS becomes unsustainable if it is instead funded by monetary expansion – both CGS and Consumer Gds Industries are competing for the same resources. This is what constitutes malinvestment.

  20. Posted 27/05/2010 at 12:52 | Permalink

    “If I invest in four machines when I only need three, then I can mothball one until another wears out”.

    This is qualitatively different from malinvestment. What you describe is day to day business decisions made with imperfect knowledge based on (among other things) interst rates set in line with market forces.

    Malinvestment is a different animal: sthg that happens on a collosal scale right across an econmomy. What causes so many entrepreneurs to make so many wrong decisions in their business plans? Answer: interest rates being out of tune with people’s time preference and, therefore, an imbalance between society’s relative spends on consumer and capital goods. Hence the collapse.

  21. Posted 27/05/2010 at 15:13 | Permalink

    If you think in your ramblings that you are investing your money with the constraints of your time. You would be on the way to understand the Austrian school.
    You will understand the kernel piont without equations.

  22. Posted 27/05/2010 at 18:13 | Permalink

    What appears as malinvestment from the perspective of efficiency can appear as a good investment from the perspective of system resilience. An economy is a complex system. The boom multiplies the number of machines per site, the number of sites per enterprise, and the number of enterprises within the local economy, the regional and the national – and, ultimately, the global. The number of linkages in the economy – supply chains in particular – are also multiplied.

    Any of these nested elements and subsystems may or may not be critical to the survival of the larger system it inhabits. If it is, then what matters is whether and how much redundancy is built into it.


  23. Posted 27/05/2010 at 18:18 | Permalink

    No business enterprise runs its plant at full capacity or without inventory. Electricity generating companies always maintain a spinning reserve, as they must ensure that supply and demand throughout the grid match exactly all of the time, and they must always stand ready to cope with surges in demand.

    Now look at the behaviour of human agents in the system. They economise, and know that it is costly to maintain redundancy. Economising behaviour tends over time to strip redundancy from the sub-systems in the economy, and once a critical sub-system loses too much of it, there is a clear and present danger of collapse of other systems which depend on it.

    In other words, bust!

  24. Posted 28/05/2010 at 15:44 | Permalink

    […] What Austrian business cycle theory does and does not claim as true, by Anthony Evans […]

  25. Posted 30/05/2010 at 18:21 | Permalink

    […] of the crisis Arquivado em: Economia,Educação,Teoria — André Azevedo Alves @ 19:20 What Austrian business cycle theory does and does not claim as true. Por Anthony J. Evans. In short, whether Austrian ideas have something to add depends on whether you […]

  26. Posted 31/05/2010 at 01:27 | Permalink


    I’ve just been abroad, looking at a factory that is the biggest of its kind in the world. It was commissioned in 2005 and closed in 2007. The owners are trying to sell it for 10% of what it cost, but in reality there is no price at which it is worth running.

    I also looked last year at a leisure facility which had gone bust within 3 years of opening. The hoped-for price was 25% of the construction cost and we valued it at less than 20%. In practice, the vendor (the lender) couldn’t afford to sell it at a rational price, because of the write-down.

    There is marvellous, mainstream, mathematical, economic theory, and then there is reality. Only the Austrians deal with reality.

  27. Posted 31/05/2010 at 12:44 | Permalink


    So, the factory you looked at has zero value except for the bare value of the site and the scrap or second-hand value of its contents. The leisure centre is worth less than 20 per cent.

    Some one is evidently going to go bust. The question is, how vital are the companies concerned for the economy and for agents at the other end of their linkages?

    Maybe Austrians deal with reality, but their problem is they have to stop as soon as they run out of conclusions necessarily contained in their premisses. After that, they can neither measure nor observe.

  28. Posted 31/05/2010 at 17:08 | Permalink

    Michael, you’ve already acknowledged that you have not taken the time to familiarise yourself with contemporary Austrian economics, so I have little patience for your assertions, but your last comment is bizarre. I’m intrigued as to why you believe that Austrians can neither measure or observe, but am hesitant to ask given how clueless your previous comments have been. Is there any substance behind that statement??

  29. Posted 31/05/2010 at 20:03 | Permalink

    There is, aje. Ludwig von Mises – A Primer by Dr Eamonn Butler states in the Summary (first bulletpoint, p13) states:

    “Economics is a science that can discover things and even make predictions – not on the basis of observation and testing, but through a process of deduction. Just as geometry or mathematics can be derived from a few simple and obvious axioms, so the science of human action can be deduced from the very concept of action and choice itself.”

    I have seldom read such a temerarious assertion. On pages 18-19 there is a skepticism about the possibility of the value of macroeconomics and its aggregates.


  30. Posted 31/05/2010 at 20:09 | Permalink


    This really will not do. It is akin to finding fault with a road map of the whole of Russia on the grounds that it doesn’t tell us the detail of the streets of Moscow, though it does tell us how to get from Moscow to Vladivostok. It all depends on the level of abstraction you want to study the subject at. I remember my very first lecture in economics as an undergraduate. We were told that you can study the subject only by making simplifying assumptions about human behaviour.

    Austrianism also has little use for mathematics (see Professor Keane’s ‘Debunking Economics, p305). That is undoubtedly a weakness, as mathematics can add rigour to analysis.

  31. Posted 31/05/2010 at 23:02 | Permalink

    Michael, I would guard against infering too much from a primer, but I don’t see the problem with Eamonn’s statement as quoted. He’s referring to the Austrian method of basing theory on deduction, which I don’t think is unreasonable. But that doesn’t deny the importance of empirical observation which any survey of Austrian work would demonstrate. I suggest you read Mises’ Theory & History to understand this.

  32. Posted 01/06/2010 at 06:21 | Permalink

    Basing theory on deduction is perfectly legitimate, provided that you can draw a testable hypothesis from it. Taken at face value, Eamonn’s statement states that you can develop economics in the same way as logic or mathematics. But economics deals with contingencies: the economic order could have been made other than it is. If you rely only on deduction, your conclusions are likely to be either trivial or wrong.

    So you have to test hypotheses against observation, and you can observe only in terms of the aggregates that summarise the economy.

  33. Posted 01/06/2010 at 09:08 | Permalink

    Hi Michael
    I think very few economists would agree that “you can observe only in terms of aggregates that summarise the economy” – economists of all schools understand that observation can take many forms. Case studies are the obvious example of how empirical work can be done without recourse to aggregation. But once more you build a strawman to suggest that Austrians deny the importance of aggregate variables. Mises and Hayek made use of aggregate variables but made important qualifications about how far to take them and what they meant. As I say, read “Theory and History” to understand what seems to be too subtle for you to grasp.

  34. Posted 01/06/2010 at 11:53 | Permalink

    Is Theory and History on the Internet to download? I have to say I haven’t heard of it. But then, when I was a student in the 1980s we did mainly neo-classical and Keynesian economics. Virtually no Austrian at all.

    I stand by my comment about aggregation, as there is simply no way of observing an economy down to the last detail. There are many different levels of it, though. Case studies are a good way of looking inductively at how a particular economy has operated at its time and in its place. Simulations are also useful.

  35. Posted 01/06/2010 at 12:46 | Permalink
  36. Posted 01/06/2010 at 12:56 | Permalink

    Hayek (he was more empirical than von Mises – as Eamonn mentions in the forthcoming video to go with the primer) talked about how you could use economic data for “pattern predictions”. They can be used to obtain a general idea of the sort of changes that might arise if a given variable changes. I think this is reasonable and accords with my experience in finance. There is an aggregate versus disaggegate problem (and you cannot assume it away, Michael, by simply aggregating at higher levels). Aggregate data can hide many things that are important; disaggregated data can clearly be partial. On this whole issue, Fatal Conceit and Hayek’s Nobel Laureate lecture are very good continued

  37. Posted 01/06/2010 at 13:00 | Permalink


    one of many problems of focusing on aggregate data is where do you stop? Is the whole world more relevant than just the UK? Presumably it must be, Michael, if the UK is more relevant than data that are just about Bournemouth. Go back to your map analogy – a map that had the roads on its from St Petersburg to Moscow might be useful; but surely it would be less useful without the more disaggregated information that they were all one-way streets, all had mines on them or were all single carriageway mud tracks not wide enough to take a car.

  38. Posted 01/06/2010 at 13:49 | Permalink

    Philip says: “Aggregate data can hide many things that are important; disaggregated data can clearly be partial.”

    Exactly the point I made earlier. A map of the whole of Russia can hide the detail of the streets of Moscow. A map of the streets of Moscow can be partial in that it might invite those who have led a sheltered life to assume that the streets of Vladivostok are similarly laid out.

  39. Posted 11/08/2015 at 14:59 | Permalink

    “It all depends on the level of abstraction you want to study the subject from.”

    It’s not the level of abstraction but the thing you are abstracting from. Austrians extract from praxeological categories rather than the less justified neoclassical models, hence the debate. Yes a map is good for navigating but in truth IS an unrealistic representation as you said. Formal models may for example be useful for predicting market movements, but they operate on an incomplete set of assumptions that may not glean the MACRO economic reality. Mises, of course, drew on praxeology.

    “Basing theory on deduction is perfectly legitimate provided you can draw a testable hypothesis from it.”

    Michael how do you justify the necessity of a testable hypothesis? This debate more or less comes down to a broader investigation into the epistemology of science itself. An empirical test of economic data necessitates an explanation. But economics appears to be in an impossible place for deriving a feasibly realistic model or theory and most are loaded with unrealistic assumptions.

    I would recommend “economic science and the Austrian method” by Hans Herman hoppe rather than theory and history and then go from there. I don’t think this settles the debate here but it would help to have understanding. Even a lot of Austrians have a hard time debating this position.

Comments are closed.