Recent Posts

Wednesday, May 10, 2017

Book Review: Can We Avoid Another Financial Crisis?

Professor Steve Keen, the well-known critic of mainstream economics and professor at Kingston University, published Can We Avoid Another Financial Crisis? His argument that unsustainable private debt dynamics make a future crisis of some sort inevitable; and the biases of mainstream economics helps ensure that little useful will be done to prevent this outcome. The book offers an overview of his economic theories, which are somewhat distinct niche within the heterodox tradition. Although the book is interesting, I have some reservations about his strategy vis-à-vis mainstream economics within the book.

Book Description

Polity Press published the book in 2017, and the paperback edition is a brief 129 pages (excluding end matter). The book consists of six chapters:
  1. From Triumph to Crisis in Economics
  2. Microeconomics, Macroeconomics, and Complexity.
  3. The Lull and the Storm
  4. The Smoking Gun of Credit
  5. The Political Economy of Private Debt
  6. A Cynic’s Conclusion
Steve Keen is the author of Debunking Economics: The Naked Emperor Dethroned?

This review is going to discuss the book roughly in the reverse order of the chapters.

Can We Avoid A Financial Crisis?

At the risk of spoiling the punch line, Keen believes that a crisis will not be avoided. He is not attempting to predict a particular date, but argues that unsustainable debt buildups will reach a head in 2020. In his view, policymakers are following poor guidance from mainstream economics, and will not make an effective response.

The only caution I would draw from this conclusion is that the logic of Keen’s models suggests that private debt buildups will reverse, causing a slowdown, but the implications for a financial crisis are less clear (as defined by defaults cascading through the financial system).

I have my home country in Canada (which Keen views as being one of the main targets of a slowdown) as an example. It is easy to see how the spectacular pop in house prices and building activity will reverse by 2020. However, it is unclear why this will turn into a financial crisis.

Obviously, weaker links in the financial system would be culled (a process that one could argue that has started), but the core of the system is protected by CMHC guarantees, and lender-of-last-resort operations if need be. If you worked at one of the smaller non-standard landers (or were an investor in their securities), it might feel like a crisis, but it is unclear why the big banks will be discomfited by smaller competitors crashing and burning.

Looking further afield, one may note that investment bankers have filled most economic portfolio appointee positions in recent years. Although this perhaps is not attractive politically, one may note this means that governments are staffed with people with the requisite background to manage bailouts of the financial system.

The Credit Engine

The driving force for Keen’s analysis is his economic model. It was improved because of contact with post-Keynesian economists, but it appears to follow the trends of the model’s earlier iterations. The model is straightforward, consisting of a few accounting identities as well as basic behavioural relations. The dynamics suggest that the driver of growth during expansions is the tendency of private sector debt to outstrip income growth. This becomes unsustainable, and then the cycle goes into reverse. However, this is the result of the reversal of borrowing in the non-financial sector, and not necessarily a disruption in the financial sector.

I have not spent much time looking at Keen’s model. I have an aversion to continuous time economic models, and I would prefer to work with the standard discrete time stock-flow consistent framework. (For those unfamiliar with the jargon, a discrete-time model has a time axis that consists of steps, such as months or quarters. A continuous time model has the time axis being the real number line, similar to most models in physics.) I believe it would be straightforward to create a model in discrete that emulates the behaviour of Keen’s model using my Python framework (for example). I expect that I will discuss why I prefer discrete time models in a later article.

One concern with Keen’s framework is that the causality is unclear: does growth cause debt, or is it a side effect of growth?

The key example is the case of housing. Most of the growth in private debt since the early 2000s in the developed world can be traced to housing markets. Since households’ down payments have at best grown in line with income, rapid house price increases imply a need for larger mortgages. We could therefore argue that high house prices cause greater debt growth. However, in practice, the higher house prices are associated with loosening credit standards. This was certainly the case in Canada; the start of the explosion in household debt was associated with the loosening of CMHC lending standards in the late 1990s.

The difference in interpretation matters. Policy makers were only able to control the lending standards; the amount of debt that results was an outcome of policy shifts. Correspondingly, it would be difficult to view private debt levels as a specific target for policy, in the same manner that the inflation rate currently is. Therefore, Steve Keen’s arguments would need further fleshing out before being put into practice.

“Modern Debt Jubilee”

Professor Keen proposed what he calls the “Modern Debt Jubilee”: the central bank transferring money to private sector accounts (similar to “helicopter money”), but with the proviso that the money be used to pay down debt first. As he notes, the implementation is more difficult than it is to describe. (Does paying bills count as “debt”?)

I am not a fan of “helicopter money” (as I discussed in essays in “Abolish Money (From Economics)!” In addition to the difficulty of tying the transfer to debt repayment, I also question the optics. Why should a debt-free rich person be able to spend the transfer on caviar and cognac, while poor people are forced to reduce the balance of their mortgages?

The Political Economy of Private Debt

Keen argues in Chapter 5 that there are perverse incentives for political leaders with respect to private debt. A leader that unleashes a bubble in the private sector is lauded as a hero as a result of the increase in growth (and rising home prices). The bubble persists until the unsustainable private debt growth reverses, and a later politician is stuck cleaning up the mess. If we had a system that monitored private debt growth (similar to the hokey public debt clocks), there might be greater accountability among politicians.

Although I am sympathetic to this, I am highly averse to unelected bodies trying to police “the truth.” We have freedom of the press for the reason; it is up to political parties to present their case themselves, without being moderated by self-selected experts.

The History of Keen’s Models

Chapters 3 and 4 discuss the history of Keen’s models since the early 1990s, and the various debt-financed booms that happened during that period. He points out that his models hinted at the various crises that afflicted developed economies, while the mainstream was proclaiming the “Great Moderation” thesis. For readers who are new to economics, this history is certainly interesting. For readers more familiar with the history, it may be that Keen’s views will be quite novel when compared to conventional histories.

The Darned Mainstream

For those familiar with Keen’s other writings, Chapters 1 and 2 cover familiar ground: mainstream economics is all wrong, and they completely misunderstand everything. For readers who are new to the topic, these chapters act as an updated summary of Debunking Economics.

I plead guilty to writing pretty much the same thing in many of my articles. However, my feeling is that Keen should have held back in this case. In this book, he is arguing that the developed world is facing a crisis that will be of a similar magnitude to the Financial Crisis. Even if mainstream economics is part of the problem, writing the book in such a way that no self-respecting mainstream economist will go past the first few pages is not exactly a great method for coalition-building behind your policy remedies.

At the minimum, this discussion should have been deferred to the middle of the book, with parts moved to an appendix. I fail to see how assumptions about the aggregation of demand functions is going to cause mainstream economists to back policies that will cause a financial crisis. Alternatively, there is little reason for anyone other than friends and family to care what Robert Lucas in particular thinks. Instead, we need to ask: is the policymaking response to a potential crisis correct or incorrect, and why?

Concluding Remarks

Steve Keen’s models and thinking are interesting, and he is certainly a passionate writer. However, my feeling is that the beginning of the book put too much emphasis on theoretical disputes, and less on the topic at hand: crises.

Can We Avoid Another Financial Crisis? (Amazon link)

(c) Brian Romanchuk 2017


  1. "I have an aversion to continuous time economic models, and I would prefer to work with the standard discrete time stock-flow consistent framework"

    I'd argue both are flawed. Discrete time models are too coarse. The fun in Steve Keen's model happens on the ground due to the difference between the intra-day mechanisms for running a financial system and the inter-day mechanisms. I don't think that is captured cleanly and in a way that is useful for policy construction by system dynamics.

    We need an agent simulation to do that.

    "Why should a debt-free rich person be able to spend the transfer on caviar and cognac,"

    That's always been the flaw I've seen in the Debt Jubilee idea. You've just equity released the rich and diverted the production system towards servicing them. How does that help equality of income - unless you are a fervent believer in trickle down theory.

    But it is another of these "no government required" ideas that economists seem strangely fond of.

    "is not exactly a great method for coalition-building "

    Neither is having more than one political party, but they exist.

    Why? Because there are fundamental disagreements on substantive philosophical normative points. The top down, higher law, arrogant, "we know best" view of mainstream economists is never going to be something you can reconcile with. It is Conservative vs. Liberal. Protestant vs. Catholic.

    You can't go with them. You'll have to go around them.

    1. Agent-based simulations are more realistic, but there are costs: the model becomes a black box. Of course, when you end up with 80 equations, it's still a black box...

  2. btw, why didn't you use Simpy in your model framework?

    1. A) Never heard of it until you mentioned it.
      B) The system builds the equations automatically. Unclear how to emulate in another package.
      C) I want control over the solution method.

  3. Keen makes an important point in his final paragraph, which is that the deflationary effect of private sector debt deleveraging can largely be avoided by Keynsian / MMT measures, i.e. having the state print money and spend it, and/or cut taxes. But he claims politicians and the economists advising them are so hopeless that they wouldn't know what to do come another serious recession.

    Keen is at least half right there.

    1. He's probably right about the advisors, but it is not just a "mainstream" issue; we have at least half the political spectrum gone unhinged.

      That said, if a crisis hits, any attachment to austerity will disappear. Politicians have better survival instincts than economists.

    2. "That said, if a crisis hits, any attachment to austerity will disappear. Politicians have better survival instincts than economists."


      You're not serious are you?

      Politicians are defined by ideology.


    3. There was a coordinated wave of fiscal stimulus in response to the global financial crisis. Too small, but it was in the right direction.

    4. The US may have reacted in this way but Europe took an alternate path, did it not?


  4. Interesting observation about the CMHC guarantee. One thing that appeals to me about the Post-Keynesian school of thought is the focus on institutional structure, particularly that of financial institutions; finance is not just a veil. I haven't read Keen's book but I do follow Keen and I wonder how relevant his research is without addressing institutional structure.

    I guess a way of looking at it would be that focusing on private debt to GDP levels doesn't take into account a major source of the financial instability which is the evaporating source of financial assets on the balance sheet of banks (or private insurers of the debt). In the case of Canada (my interpretation of you description), the balance sheets of banks aren't under thread because they are guaranteed by the CMHC. Similarly I wonder if the China will have a crisis given that a lost of private debt is owed to state owned banks that the government would recapitalize in a crisis.

    1. Keen does discuss the situation in China with the institutional structure in mind. So he's not completely ignoring institutions. However, this is a complication for his call for a focus on private debt; once we start arguing about institutional detail, who's to say that any particular debt growth rate is unsustainable? Perhaps it is just the economy moving to a new steady state with a higher "normal" debt level.

  5. This comment has been removed by a blog administrator.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.