The case for eclecticism


Most macroeconomists have a mental model of how the economy works. I think there are five relevant mental frameworks:

  1. Keynesian. This is the approach that most people are taught in high school and undergraduate economics, is believed by most economic policymakers, and works most of the time. The economic cycle is caused by fluctuations in optimism and pessimism, and the policy response is to lean against the wind with fiscal and monetary policies. ‘New Keynesian’ economics is a subset of this thinking, which – largely by assumption – places all the heavy lifting on wage and price stickiness, and integrates rational expectations – not for realism, but for mathematically sophisticated (and psychological crude) modelling of expectations. In reality, the financial sector is (still) very simplistically represented in the Keynesian mainstream. Balance sheets are reduced to net wealth. So there is a household net wealth effect. Interest rates affect demand simplistically, through inter-temporal substitution and investment spending. Since the financial crisis, the financial system is now bolted on, but there are no agreed conventions. For example, most mainstream macro Keynesians don’t really have a framework for thinking about equity risk premia, credit spreads, and the determinants of lending conditions and other market interest rates – hence all the hoo-ha over the recent trivial move in interest rates by the Federal Reserve, epitomised by this fascinating debate between DeLong, Summers and Krugman.

  2. Classical theory. Classical theory assumes that all markets clear – it’s an assumption, because we all know they don’t. As a result, there are no cycles, and growth is determined by productivity and labour supply. The determinants of productivity are more contentious as are the determinants of capital accumulation. Most macroeconomists when they think about long term growth are classical economists. Real business cycle theory is usually classed as ‘classical’, but its innovation is to identify technological shocks as a source of cyclical variability – which is periodically relevant, but can be just as easily integrated into a Keynesian, or classical, view. Liberated from its ‘classical’ context, the idea that technological change causes errors of extrapolation and cyclical disturbances (particularly sectorally) is quite useful.

  3. Minskyian theory. The orthodox-heterodox divide is largely about academic turf wars and the university syllabus than real-world economics. Most economists believe some, if not a lot, of Minsky.[1]

    Tobin and Samuelson, Keynesian economists who thought a lot about finance, were also very influenced by Minsky, even if this is not evident in their theoretical work (particularly in Samuelson’s case). Minskyian theory puts the financial sector, household and corporate balance sheets at the heart of macro thinking. In essence it is the most rigorous cyclical and behavioural macro theory of asset prices and leverage. Stock prices and other asset prices are seen as cyclical, and periodically euphoric – leverage similarly follows cyclical ‘stages’ based on psychologically-biased beliefs about asset prices and the sustainability of growth. The financial system is inherently cyclical and perniciously destabilising. Tony Yates makes a diligent attempt to show that the Keynesian mainstream doesn’t need Minsky. Steve Keen  makes clear what it’s still missing. Simon Wren-Lewis is typically balanced and perceptive on orthodox and heterodox disputes.[2]

  4. Rational expectations. This should be thought of as a macro framework in its own right. The premise is that the most relevant way to understand macroeconomic effects is to assume that policy measures are neutralised by the private sector pre-empting policy-makers intentions. Like with real business cycle theory, the association of this framework with market-clearing dogmaticism is an unhelpful barrier to thought.  Ironically, rational expectations theory is most relevant to economies with less competitive micro-economic structures – for example emerging markets which have institutionalised indexation. This is why ‘rational expectations’ was a very valid critique of policy-making in the 1970s. In a world of centralised and collectively determined prices and wages, it is highly likely that attempts to use inflation to ‘trick’ trade unions or monopolistic firms will fail. But in a world of price stability, as Greenspan defined it – i.e. low enough aggregate inflation that private agents ignore aggregate price forecasts in their decision-making – ‘expectations’ are no longer relevant to macroeconomic outcomes.

  5. Marxian theory. Although economists are notorious for modelling individuals as self-interested, most macroeconomists ignore the likelihood that groups also act in their self-interest. Marxian theory also has insights into the distributional and social consequences of technological change.  Marx recognised that technology determines the way production is organised, which in turn determines the social structure and distribution of income – so most of the current analysis of the effects of technology is essentially Marxian. As is the view that ‘austerity’ serves vested interests.[3]

These descriptions should already reveal my preference for eclecticism. Freed from advocates’   desire for all encompassing theories, and biased ideology, I think there are profoundly useful insights from each perspective – and all perspectives are needed to understand the world’s current macroeconomic picture.

For example, to understand what is happening in China, you have to use Minsky – as Krugman’s superficial analysis makes clear. The main risk is in the financial sector. The post-2008 macro stimulus in China was hugely successful as a short-run boost to domestic demand, but it was effected through rapid house price inflation, and growth in corporate and quasi-governmental leverage. Leverage errors based on growth and asset price extrapolation are profoundly dangerous and difficult to fix.

China has some advantages. There is some truth in a classical perspective that leverage and asset values are simply book entries which can be wiped clean at the stroke of a pen – the problem is whose pen, and in whose interest. At least the Chinese government already owns most of the banking system – so the coordination of debt restructuring may be easier, for similar reasons the local government debt overhang may prove more tractable than our private sector, or the Eurozone’s quasi-sovereign, variant.

But to understand why global price stability has been resilient to all sorts of shocks in the last 20 years, Minsky is useless (as his inflation forecasting record suggests). You have to think like an old-fashioned classical economist – it’s because of deregulation and competition. Sorry central bankers – you were a sideshow (the proof is that you can’t raise inflation ‘expectations’ now that you’re trying … and if you can’t cause inflation to rise – you probably didn’t cause it to fall).

In order to understand why emerging markets raise interest rates and have recessions when their currencies weaken, you have to understand Lucas and rational expectations (just remember that the term ‘rational’ can be correctly applied to institutions not individuals). Many emerging market economies have 1970s institutional structures (and – a separate issue, significant US dollar debt). Trade unions in Brazil and South Africa are entirely rational in raising wage demands in response to devaluation – although collectively it may be self-defeating. So in macroeconomics the distinction between emerging and developed markets is really about microeconomic structure and borrowing in foreign currency.  As a result, countercyclical policies are severely hampered, and exchange rate weakness is not a straightforward ‘stimulus’.

By  contrast, I think the US economy currently fits a conventional Keynesian picture: the last recession, followed by the Eurocrisis, some silly fiscal squabbles and a prolonged hangover from the housing boom, are severe blows to animal spirits, which are gradually strengthening as memories fade. Ditto the UK. To this extent Paul Krugman is spot on.

As for Europe, classical growth theory gives the main insight – it’s growth plight is primarily demographic – you need to think about Solow.

Perhaps the biggest gap in our cannon is how to explain global real interest rate trends. This may require a synthesis of classical growth theory, Minsky, and some eclectic old-Keynes. Shifts in savings preferences caused by demography, rising per capita income and perhaps the distribution of income are plausible classical explanations. But there is a behavioural, path-dependent overlay which is likely relevant. When real interest rates peaked in the 1980s there was a very high embedded inflation risk premium. In this sense, policy rates (and bond yields) were not ‘risk free’ rates. Furthermore, the dependency of repeated subsequent policy stimulus on interest rates encouraged household leverage. To some extent, further stimulus required ever-lower rates, as households operate with simple interest-affordability rules of thumb. Running current account surpluses as a means of financial insurance in Asia against the volatility of global capital flows also exacerbates these trends.

Perhaps the greatest irony in macroeconomics is that the world’s main macro problem – insufficient demand – is an area where all frameworks are in agreement as to what should be done. Helicopter drops of cash to households raises spending in Keynesian, monetarist-classical, and Minskyian models. But no one wants to do it … Ultimately, evolutionary models are most relevant: why else would it be better to fail conventionally?

[1] I happened to mention MMT to Ben Bernanke at a dinner in late 2014. After explaining to him the basic tenets, he replied: ‘Isn’t that just economics?’

[2] Since writing this blog, Jan Zemanek aka @wonkmonk has drawn my attention to Stiglitz’s integration of Minsky in recent work. Roger Farmer also stands out for rigorous and plausible integration of asset prices and a “Keynesian” perspective which does not rely on sticky prices.

[3] Chris Dillow exemplifies how pertinent Marxian analysis remains.

About The Author

Eric Lonergan is a macro hedge fund manager, economist, and writer. His most recent book is Supercharge Me, co-authored with Corinne Sawers. He is also author of the international bestseller, Angrynomics, co-written with Mark Blyth, and published by Agenda. It was listed on the Financial Times must reads for Summer 2020. Prior to Angrynomics, he has written Money (2nd ed) published by Routledge. He has written for Foreign AffairsThe Financial Times, and The Economist. He also advises governments and policymakers. He first advocated expanding the tools of central banks to including cash transfers to households in the Financial Times in 2002. In December 2008, he advocated the policy as the most efficient way out of recession post-financial crisis, contributing to a growing debate over the need for ‘helicopter money’.

5 Responses

  1. David Harold Chester

    With so many models floating around, it surely is necessary to determine a criterion for choosing which is best suited to understanding the problems in macroeconomics. In my view, and this should be a theorem, that unless the model represents the whole of our social system it is inapplicable. This is because the results of partial models are so different and one cannot trust them and their (often unstated) assumptions.

    I would go further and claim that a good model (like Prof. Einstein’s criterion for a good scientific theory) should be made as simple as possible without being over-simplified. On this matter I have done some research and the model that I have produced is able to meet these requirements and it enables us to better understand of what our social consists and about how it works. Not many summary models of this kind exist, but they have use because one can then expand them in the places of interest and avoid that irksome requirement of “all other things not being allowed to change”.

    The relatively simple model that I use may be found in my paper SSRN 2600103 “A Mechanical Model for Teaching Macroeconomics”, which also refers to my book on this general subject.


Leave a Reply

Your email address will not be published.

* Checkbox GDPR is required


I agree