<![CDATA[pseudonotes: Theory]]>https://www.pseudonotes.com/s/theoryhttps://substackcdn.com/image/fetch/$s_!hdUg!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb0f465e8-5cc1-4fc5-95ea-c119f86995bc_512x512.pngpseudonotes: Theoryhttps://www.pseudonotes.com/s/theorySubstackFri, 25 Sep 2026 21:10:41 GMT<![CDATA[Every Recession Is a Policy Choice]]>https://www.pseudonotes.com/p/every-recession-is-a-policy-choicehttps://www.pseudonotes.com/p/every-recession-is-a-policy-choiceThu, 10 Sep 2026 15:05:04 GMTalmost all economists agree that after a boom, a bust is in some sense necessary. some call it a ‘necessary correction’, others overaccumulation. even post-keynesian and institutionalist thinkers tend to hold on to some version of the idea – that discipline is needed, that imbalances must be resolved, that without periodic stress the system drifts further from health. i think this is wrong. not wrong in emphasis, but wrong at the root. what follows is an attempt to show why, by reconstructing what booms actually are, what busts actually do, and what is really at stake when governments let busts become recessions.

this matters less because of lost output and more because of the social consequences of recessions:

The proportion of people without secure access to food, housing and other necessities is a much better measure of the economic costs of recession than the fall in GDP. But even this is the smaller part of the cost of unemployment. The most important thing about work, under capitalism, isn’t that it produces goods and provides an income, but that it is the carrier of self-worth, status and social power. (...) Persistent unemployment breaks social bonds, with profound effects that don’t show up in the aggregates.
J. W. Mason (2010)

*

this text is written as compressed theses, not in fully developed argument. each thesis could be – and in many cases hopefully will become – a longer piece. but the point of presenting them together is that they belong together: the nature of booms, the mechanics of busts, the question of whether busts are useful in any sense, and the political economy that treats them as inevitable have to form one argument, not four. separating them is how so much of the existing literature has produced many astonishing parts that then never add up to an informing whole. i think we have enough of that. i would rather try to be specific enough to be wrong in informative ways than careful enough to be right about yet another aspect.

boom

how growth works

  1. growth is driven by capacity expansion, which itself is motivated by strong demand outlooks. it therefore relies on one or more drivers: strong technological stories or government programs creating capex in some parts of the economy, which then fuel a larger boom. the availability of funding plays no fundamental role in this. (see Keynes 1936)

  2. capacity expansion is imbalanced – with some sectors expanding way beyond their profits, taking the bets that create income and jobs for the rest onto their balance sheets. the money balances generated in that credit-driven process are funding exactly this imbalance. booms are always uneven and combined development, they are never just a broad expansion of everything all at once – not even in highly state-led booms. (see Kalecki 1954) these imbalances are also rarely confined to a nation state – typically, they aren’t even scoped to an economic region (defined by a common currency or stable FX on the one hand and legal/political stability on the other). rather, they are global. (see Hirschman)

  3. the imbalances within booms can be more or less pronounced. the way they become less pronounced are twofold:

    1. when labor compensation goes up – not only in some specializations but broadly. this makes the boom as stable as growth in a monetary production economy gets because it creates income flows that then increase consumption – therefore decreasing the reliance of the overall boom on further balance sheet expanding bets. (on Wage-Led Growth, see Bhaduri/Marglin and Baccaro/Pontusson)

    2. when it becomes easier for households to get mortgages. this is not as stable because it indebts actors with little equity and no easy means of default based on asset prices that are prone to become speculative. it is also highly dependent on institutions: financial engineering and homogenisation as well as government derisking. these institutional changes drive growth of this more than the income growth and income stability of the debtors – the ‘fundamentals’ if you will. (see Mian/Sufi and Jordà/Schularick/Taylor)

  4. the real imbalances correspond to financial ones: the ones willing to expand their balance sheet and take risk need to be matched to those who get additional income from working for these directly or indirectly – building factories, houses, semiconductor chips. that’s where finance comes in. it creates the chains of IOU necessary to meet the preferences of the owners of the additional money balances with something that funds the boom. for example MMF → SIV → SPV → mortgage. (see Mehrling and Pozsar)

  5. because of this, aggregates are never the constraint. “funding” is never the thing that restricts growth as such . although, funding conditions play a major role in (a) how the boom is funded and also (b) which parts of the economy are booming at what pace. (see Tankus) any boom can be financed – the question is just, which liquidity and solvency risks are created in the process and which balance sheets are taking what part. that’s the dynamic web of interlocking balance sheets, and it is naturally hard to study. (see Minsky, Godley/Lavoie. for cross-country: Brender/Pisani)

  6. therefore, aggregates also rarely tell good stories. this applies to monetary aggregates as well as to statistics on corporations or households.

bust

what can go wrong depends less on what the boom built and more on how it was financed

liquidity crisis: somewhere in the funding chains, a creditor refuses to roll over funding. there are always other players able to step in and provide the funding – the question is whether they are willing to do so. when uncertainty about booming trends is high or the exposure of certain parts of the financial system to them is opaque, this might not be the case. also, this can break because of some external factor creating uncertainty in such a way that everyone wants to be more liquid (covid). the central bank can always end liquidity crises: “lend freely, at a penalty rate, against good collateral” (Bagehot)

examples: GFC 2007-2008, Covid 2020

a boom story collapses. this is typically why the stock market crashes. the demand generated by the capex involved will go down and reduce income. this can be offset by government programs. other effects depend on how the boom in these sectors was funded: if truly funded by equity, it is just a loss for investors. if investors financed their equity leveraged though, it can become a liquidity crisis as well as a balance sheet recession (see below). and if enough of it was financed by the financial system, it can also create financial system insolvency (see below).

examples: Dotcom 2000-2001, Railroads 1873 and 1893, Fracking 2014-2016, Maybe AI soon...

a real estate bubble pops: firstly, just as with any boom story, the economic activity driven by the construction of new real estate goes bust. again, this can be offset by government programs. for the funding side, the structure is typically different, with solvency risk spread broadly. this is why this typically creates a...

examples: US Subprime 2007-2009, Spain and Ireland 2008-2012

balance sheet recession: household balance sheets might be under water, forcing them to severely reduce spending on other things to get their balance sheet more comfortable again. corporate balance sheets might be under water as well (this is what made the japanese case of a balance sheet recession so deep). both cases can be problematic because they threaten the social contract. as long as the further demand reduction is truly offset by government programs, it needn’t be a bigger problem. (See Fisher, Koo)

examples: Japan 1990-2003, GFC 2008-2012

financial system insolvency: if the solvency risk taken by the financial system exceeds its equity plus its realistic short term profits, it has to be bailed out. economically, there isn’t really any issue with this, and time and time again, governments who did this made a profit from it. politically, it is often contested, as for example in the japan case, which was prolonged – beside other reasons – by an unwillingness to bail the banks out.

examples: Japan 1997-2003, Sweden 1991-1992

another thing that can happen is a price shock and in some cases, conflict inflation – but this is not our issue today. as an introduction to that topic, see Eich/Tooze and for the more recent shocks, Weber/Wasner. inflation might actually be challenging for governments – because there are real tradeoffs involved, not only the immediate distributional ones but also structural questions of power (Kalecki 1943). recessions on the other hand are economically trivial to solve in theory, however, it might be limited not only by politics, but also by state capacity. nevertheless, in that sense, every recession and every following stagnation is a policy failure.

why recessions happen

  1. the liberal version of political economy is to treat the economy as something that naturally and endogenously creates growth with the state merely being responsible for the regulatory and infrastructural framework. this was not invented by thatcher et al, but deepened.

  2. politically, this comfortably allows for almost anything to be publicly discussed as being a “market outcome”. regions prospering is just as much a market outcome than regions in decline, mass unemployment and deaths of despair.

  3. during booms, consensus holds the central bank responsible for price stability and banking regulation and the fiscal state for education, infrastructure, energy and labor market design.

  4. during busts, politics is focussed on finding the regulatory framework mistakes much more than on what is still known as countercyclical policy. if liberal states didn’t have the automatic stabilizers created in a politically different, much more keynesian era, these crises would be much worse. nevertheless, growth doesn’t automatically reappear. with growth stalling and the CB setting rates to zero, parliaments and public discourse debate what it is that the economy is lacking: some blame bureaucracy, the others labor laws and the less liberal voices international competition. consequently, demand continues to be depressed until some technological story or some external shock revives the economy.

  5. meanwhile, china is showing the world what many “developed” countries never truly believed, moreover actively try to unlearn for over 50 years now:

Anything we can actually do, we can afford.
Keynes (1942)

on discipline as medicine

no doubt, discipline for individual corporate units probably is a healthy thing in a system of monetary production. if units sometimes feel the survival constraint of the market, that is useful discipline in the sense that it ensures efficient resource use – setting some or all of these resources free for other uses. that is the survival constraint, delivered by competition – no need for a recession.

this also applies to banks.

for households and humans in their role as resources of corporations, losing one’s job is a tragedy. there is also, cynically speaking, nothing efficient about job loss. while elasticity of the labor market is important in the sense of people being able to move jobs in reaction to a shift in demand and investment patterns, this does not constitute a need for unemployment. when there is something more important to do, the responsible units of production simply have to provide more attractive working conditions (typically: higher pay).

what is neither necessary nor useful for any economic goal is for the household sector or the corporate sector as a whole to be under stress. aggregate discipline serves no apparent purpose and creates certain misery. no firm needs a recession to fail, and no worker needs a recession to change jobs. recessions are not discipline. they are the absence of policy.

References

Notes on the Crises
Low Interest Rates Don't Drive Market Concentration
Dear Readers…
Read more
6 years ago · 48 likes · 9 comments · Notes on the Crises
]]>