Home > Uncategorized > Thomas Sargent discovered his inner Marxist. Really. Two graphs.

Thomas Sargent discovered his inner Marxist. Really. Two graphs.


Graph 1. Unemployment in the USA, % of the labor force, monthly data.

One of the central and most pressing questions of macro-economics is how to estimate and explain unemployment. Thomas Sargent, card-carrying member of the neoclassical cabal and winner of the ‘Sveriges Riksbank Prize In Economic Sciences In Memory Of Alfred Nobel’ (SRPIESIMOAN)  just made a shot at it. A somewhat Marxist shot, as far as I’m concerned. Which, considering the hard core neoclassical nature of the rest of the work of Sargent, is quite surprising. What’s the case?

Unemployment as we measure it is a cyclical variable (graph 1). The increases of unemployment coincide perfectly with the downswings of the business cycle which are measured using methods pioneered by Wesley Mitchell (the grey bars in graph 1). Also, declines take much more time than the increases. The cyclicality of unemployment is a problem for neoclassical macro-economics. Neoclassical macro has a binary view of human time: you’re either at work in the market (pain) or you’re enjoying leisure (pleasure). Unpaid household work or not having a job but seeking one are as a (common) rule not acknowledged as separate activities. Yeah, there are exceptions to this rule. But your typical neoclassical macro ‘DSGE’ model does not incorporate such exceptions. Which means that the models can’t explain unemployment simply because unemployment is not part of the models. ‘Unemployment is leisure’. I’m not kidding. Another recipient of the SRPIESIMOAN, Robert Lucas, as well as Edward Prescott (a winner too) and Mortensen and Pissarides (also SRPIESIMOAN winners) often or, in the case of Prescott, always denote unemployment with the word: ‘leisure’. This modeling strategy  of course does not make unemployment go away. And it doesn’t make it less cyclical. Or less important. Or less scarring, scaring and depression inducing. And especially after 2008 it became urgent to adapt the neoclassical model to the variable that could not be ignored anymore. Thomas Sargent (together with Lars Ljundqvist) tried to make sense of it all. Reading their work, it reminded me of something I had read not too long ago, ‘The processes involved in business cycles’ by Wesley Mitchell (1927). About Marx, Mitchell writes:

The most vigorous attempt to prove that crises are a chronic disease of capitalism, however, was that made by Rodbertus and elaborated by Karl Marx. The germ of this theory also is found in Sismondi and Robert Owen. Wages form but a fraction of the value of the product and increase less rapidly than power to produce. Since the masses dependent upon wages constitute the bulk of the population, it follows that consumers’ demand. cannot keep pace with current supply in seasons when factories are running at full blast. Meanwhile the capitalist-employers are investing their current savings in new productive enterprises, which presently add their quotas to the goods seeking sale. This process of over-stocking the market runs cumulatively until the time comes when the patent impossibility of selling goods at a profit, or even at cost, brings on a crisis.

Now,Sargent and Ljundqvist):

For any model with a matching function, to arrive at the fundamental surplus take the output of a job, then deduct the sum of the value of leisure [he means wage costs, M.K.] , the annuitized values of layoff costs and training costs [labor costs too, in modern accounting, M.K.] and a worker’s ability to exploit a firm’s cost of delay under alternating-offer wage bargaining, and any other items that must be set aside. The fundamental surplus is an upper bound on what the “invisible hand” could allocate to vacancy creation [Mitchell calls this ‘the current savings of the capitalist-employers’, M.K.]. If that fundamental surplus constitutes a small fraction of a job’s output, it means that a given change in productivity translates into a much larger percentage change in the fundamental surplus [duhhh.., M.K.] . Because such large movements in the amount of resources that could potentially be used for vacancy creation cannot be offset by the invisible hand, significant variations in market tightness ensue [‘the patent impossibility of selling goods at a profit, or even at costs’, M.K.], causing large movements in unemployment [‘brings on a crisis’, M.K.].

Graph 2.

Gross creation

The ‘Matching functions’ mentioined in the quote explain unemployment by assuming that finding a job or a worker takes time. And this does explain unemployment – part of it (2%-point?). The rest must be explained by crises and the inability of the market system to create jobs. As is clear from comparing graph 2, short-lived crises cause lower levels of job creation and higher levels of job destruction. Basically, these swings are not even that large. But together they lead to a fast increase in unemployment which take years to overcome. Sargent and Ljundqvist did re-invent the wheel. If they had red Rodbertus, Sismondi, Marx, Owen or Mitchell they would have known.

Fun fact: the neoclassical ‘DSGE’ model of Bokan e.a. distinguishes a class of bankers, a class of entrepreneurs (let’s call them ‘capitalists’, as they own all the capital) and a class of households which have nothing else to sell than their labour… The model knows a ‘positive wage mark-up’ but change this into a ‘wage mark down’ (for instance caused by ‘monpsonie’ on the labor market, i.e. by strong labor  market power of employers, and it’s starting to look pretty Marxist, too.


  1. November 12, 2018 at 10:34 pm

    Are you saying the Chicago school of economic thought is bellwether neoMarxist? It’s true yet funny to say it like that.

  2. November 12, 2018 at 11:09 pm

    Analysis and subject choice are both brilliant. The opening; exquisite.

    ‘Sveriges Riksbank Prize In Economic Sciences In Memory Of Alfred Nobel’

    More commonly known as; SRPIESIMOAN

    [the r is silent]

  3. Craig
    November 13, 2018 at 12:30 am

    How many times do we have to beat the dead steed of DSGE before we start crafting actually resolving policies that replace it???

  4. Helen Sakho
    November 13, 2018 at 2:28 am

    So, let us pose a question to all, especially the younger Economists:
    1) Explain, without justification, where the inner and outer of any analysis coincidence?
    2) Explain the time lag involved.
    3) Marx is dead. Marxism died with the death of international solidarity and all that went with it.
    True or False? Analyse critically.
    4) Explain, as we asked you to too many times before (sorry for the boredom that might make you fall asleep) when will totally avoidable famines , wars , and conflicts stop, even if high rates of profits are legally welcomed?

  5. Frank Salter
    November 13, 2018 at 10:38 am

    It is only possible to analyse unemployment and economic cycles — they are two views of the same thing — by analysis of transients. The differential equations need to be expressed as changes in time. Equilibrium theory is an oxymoron.

    However, it is not necessary to provide a quantitative solution. A qualitative analysis is sufficient to reveal the policy implications. A lack of demand triggers rising unemployment. Lack of demand arises from too low a labour share. Increasing the minimum wage substantially will reduce unemployment — well established empirical fact which differs from conventional analysis.

    • Frank Salter
      November 13, 2018 at 10:42 am

      PS, I should add that trade unions should be encouraged to improve the labour share. It is the power of employers to restrict workers pay which finally triggers the next economic collapse.

  6. patrick newman
    November 13, 2018 at 11:34 am

    The true measure of unemployment would be a regular survey of those of working age who are not registered as unemployed but are seeking work or would like/need to work plus those registered. This would be superior to declaring millions as ‘economically inactive’. Furthermore, you could reasonably include a measure of underemployment – those working less than a full standard week but would like to work more hours!

  7. I say!
    November 15, 2018 at 5:53 pm

    Lies, damned lies, and statistics

    Often quoted “official unemployment number” is not a measure of true unemployment, because it does not count all people.

    This should be obvious to all intelligent observers as you look at masses of people who “join the workforce” in both Japan and US as demand for labor grows

    It’s just a statistical trick to fool people – and we should question it’s use everywhere we can. And demand TRUE measurement of actual unemployment.

    • merijntknibbe
      November 17, 2018 at 11:13 pm

      This sure is a problem. The BLS probably should add discouraged workers and people marginally attached to the labor force (both categories are measured) to the labor force.

  8. November 23, 2018 at 7:50 am

    As they say the blame or credit, depending on your view point for the invention of the “unemployment rate” is distinctly non-theoretical. Rather it is thoroughly political and pragmatic. Until the early twentieth century, significant economic hardship was so woven into human history and experience that it wasn’t yet seen as an aberration. Only the combination of a severe crisis and the belief that such crises not only could but should be prevented by collective government action led to the creation of unemployment statistics, the latter created by the Progressive Era of the early twentieth century. That then led to the demand for statistics for prices, production, and industrial output and for the entire panoply of modern indicators that have come to define our twenty-first-century world. One of these is the “unemployment rate.” After 1929, President Hoover’s response to the economic crisis was like that of Republicans to previous downturns. “Batten down the hatches” and wait out the storm until the system righted itself. During this period, aside from Progressivism statistics had arisen in the nineteenth century as part of the ongoing scientific/industrial insurgency. This stirred the belief that quantifying the world could allow society to craft systems and institutions that would lead to more prosperity and power and eliminate the chaos and insecurity so endemic in human history. Statistics, the transformation of raw data into consistent numbers, are “an authoritative way to describe social problems,” and that, in turn, is a necessary first step in solving those problems. Hoover would have none of it. FDR embraced the new sciences even before entering the Presidency. The statistics on unemployment showed more than 20% of Americans had lost their jobs by 1932, not counting the thousands of farmers who lost their farms and the millions of seasonal/itinerant workers no longer working. After his landslide victory over Hoover in 1932, his administration helped begin the Bureau of Labor Statistics. With the help of those such as Senator Robert Wagner of NY who introduced and finally got passed bills that would implement the new science of statistics in government. To measure accurately not just unemployment, but also many other economic “indicators.” Also, the BLS provided the information to show FDR’s new action plans worked, put Americans back to work, and improved the American economy overall. With this Americans knew not only their destination but how to get there. It has taken five Republican Presidents over 70 years to destroy these arrangements and rob Americans of their ability to exercise control over their own futures on the job and in the voting booth. Republican Presidents from Nixon to Trump continue to mimic Hoover – Americans cannot plan or legislate themselves out of bad economic times. We can only batten down the hatches and wait out the storm until the system rights itself. And each of these Republican Presidents had manipulated BLS statistics to favor their policies. Several so much as to carry out the adage – “lies, damn lies, and statistics.”

  1. No trackbacks yet.

Leave a Reply to I say! Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.