Why labour markets don't clear

This post originally appeared on Pieria in July 2014. 

Roger Farmer has a blogpost in which he shows that labour markets don’t clear. Specifically, employment varies with the business cycle, whereas the labour force participation rate and hours worked only show long-term secular trends. During cyclical downturns, therefore, we must conclude that there is more labour available than there are jobs.

New Keynesians say that the reason for this is sticky wages. If only nominal wages could fall enough,the market would clear and there would be no cyclical increase in unemployment. Therefore there should be labour market deregulation so that wages can flex with the business cycle. Roger Farmer questions this: he argues that the market simply does not clear at any wage.

I disagree. I think the market does clear – when wages fall to starvation level. Humans need a minimum income to sustain life, but employers have no responsibility for ensuring that the remuneration of employees meets that minimum subsistence level. Therefore, in an economic downturn, the market-clearing price of labour can fall to below the minimum needed to sustain life. When wages are at starvation level, hours worked, labour force participation rate and workforce size all decline as people become weak, ill and eventually die – or, if they can, leave for somewhere more prosperous. Reducing the size of the workforce means that the market will eventually clear and wages start to rise again – for those who have survived.

This is the fundamental flaw in the “sticky wages” argument. In an economic downturn, the labour market cannot clear without incurring unacceptable social costs. Malnutrition, starvation, disease and death are the consequences of freely falling wages in an economic downturn. The reason why labour markets don’t clear is that we don’t want them to. Of course, if for some reason we think allowing people to starve to death (or emigrate) is acceptable, then the labour market will eventually clear - as it did in 19th-century Ireland. It cost the lives of an estimated one million people.

It all starts with unemployment benefits. Because we don’t like people to starve in cyclical downturns, governments in developed countries have created financial safety nets for the unemployed. Now, extreme free-market aficionados criticise these for causing unemployment: they believe that if unemployment benefits were cut or eliminated, the unemployed would be forced to find work. And they have a point. When there are unemployment benefits, people will prefer unemployment to accepting work at below unemployment benefit level. Unemployment benefits therefore act as a wage floor, preventing wages from falling below subsistence level. The result is that the labour market fails to clear and we have stubborn unemployment.

Of course, this assumes that people can choose to receive unemployment benefits rather than accept work at very low wages. But this is unpopular with the working people whose taxes pay the unemployment benefits. Working people don’t like seeing the unemployed being (as they see it) paid not to work: it seems unfair that they should work hard for their money while others get “something for nothing”. Some governments respond to this by introducing some form of workfare, where people “earn” their unemployment benefits by doing some kind of community service or unpaid “work experience”. Others impose sanctions such as loss of benefits for refusing work. And others cut unemployment benefits in an attempt to “make work pay”. The effect of this is to reduce or remove the wage floor. Without other government intervention, sanctions and/or workfare cause wages to fall to the market-clearing price.

But when wages fall below subsistence level there is pressure on government to top them up: hard-working people really don’t like not earning enough to live on. So the next stage of intervention is the introduction of in-work benefits to support wages. These place further downwards pressure on wages, since employers have an incentive to pay below subsistence level knowing that government will top up wages, and employees will accept below-subsistence wages for the same reason. When unemployment benefits cannot act as a wage floor AND there are in-work benefits topping up wages to subsistence level, the market-clearing price of labour is zero or below (an unpaid internship with no expenses is in effect a negative-wage job). This may be even lower than the market-clearing price would have been without the combination of in-work benefits with workfare or sanctions for the unemployed.
The tendency of wages to fall when there are in-work benefits and unemployment benefits are stigmatized makes a minimum wage necessary to prevent government becoming the de facto employer of large parts of the workforce. But like any wage floor, this also prevents the labour market clearing.

There is no doubt that a minimum wage distorts the labour market. These charts from the Resolution Foundation show the effect on the UK labour market of the minimum wage introduced in 1997:

(the spike at 30 on the 2012 chart is not significant – it is caused by compressing the “tail” into a single point to avoid having to extend the x axis).

There are two things to note here. First is the obvious spike at the minimum wage itself, which is much larger than we would expect from the truncation of the lower distribution. Second is the flattening of the curve. These two features are related. As the Resolution Foundation explains, the minimum wage has become the “going rate” in many sectors. It is acting as a wage ceiling as well as a wage floor. The Resolution Foundation’s report goes on to examine ways in which a surrogate wage distribution could be created to reduce or eliminate clustering at the minimum wage. Intervention in markets causes distortions which require more intervention…..In the medical profession they call this the “spiral of intervention”. The same thing happens in policymaking.

The spiral of intervention is not without costs. The Resolution Foundation’s second chart is unfortunately from 2012, which was during a period of poor economic performance for the UK. But it does suggest that the market-clearing price of labour is far below the minimum wage. I suspect that this is the case even in normal times – and it is because of the distorting effect of the UK’s combination of unemployment-benefits-with-sanctions and in-work benefits.

So wage subsidies in economic downturns are reasonable countercyclical policy, supporting people so that they can live even when the market-clearing price of labour has fallen way below subsistence level. But use of wage subsidies in normal times to support living standards because wages are persistently below socially acceptable levels is a different matter. It is systematic intervention in labour markets for political, social and moral reasons. And it raises serious questions about the nature of “private” enterprise. If labour costs in “private” enterprises have to be heavily subsidised in order to maintain an acceptable level of employment and living standards, can they really be said to be private?

Related reading:

A Very British Disease
Productivity and Employment: A Cautionary Tale
The changing nature of work
Bifurcation in the labour market
Economic equivalence: job guarantee and basic income

Image is of the Famine Memorial in Dublin. 


  1. Great insight as always. I am a big fan of your writings.

    However it would be great to hear your comment on this line - "When there are unemployment benefits, people will prefer unemployment to accepting work at below unemployment benefit level."
    I know nothing about economics but in Australia, I have seen people not accepting jobs which are well above subsistence level. Its as if, they wont take a job worth 50000$/annum in fear of losing 30,000$/annum worth of unemployment benefits.
    I am guessing this big chasm is probably the "income tax fear" or something else.

    Eager to listen to your thought.

  2. I think Keynes said that labour markets don’t clear for the following reason. Assume there’s excess unemployment and that wages fall by X%. Assume there’s excess unemployment among entrepreneurs as well and that their wage (aka profit) falls by X% as well. That means that prices will ultimately fall by X% because labour (including the labour of entrepreneurs) is the ultimate cost, and that means everyone is back where they started!

    Note that wages do not fall in REAL TERMS because wages have fallen by the same amount as prices.

    Then Arthur Pigou came along and said (I paraphrase) “hang on mate: if prices fall by X%, that means that the real value of money (base money in particular) will rise by X%, as indeed will the real value of government debt (which is a private sector asset), and that will encourage extra private sector spending in both nominal and real terms. Ergo demand rises and unemployment falls.”

    That’s the so called “Pigou effect”.

  3. Your final point is a valid one. And it's the main reason why George Osborne, when Chancellor, lost patience with "free rider" businesses looking to take a subsidy from the state. His rationale for an increased Minimum wage (the so called "national living wage") was primarily to force employers to shoulder more of their own wage costs. And it's interesting in that context to note that he did not undertake any of the detailed assessment of "what the market will bear" associated with the initiation of the original Minimum Wage - it was a blunt political assessment. Which turns our to be shrewdly judged in that there was no subsequent jobs crash. Which lends credence to the proposition that a further increase to the Real LW level might also be sustainable.

  4. It seems to me that the key issue is globalisation, if a private business in an economy with reasonable social protection has to compete with a state run slave operation in an autocratic dictatorship abroad then there is plenty of margin for an importer to undercut said private business. In the end; for many reasons and certainly environmental protection and social well-being, globalisation has to be regulated and that should be the job of national governments responding to electors needs, not lobbyists for off shore corporations. Given that not all people have the same interests, intellect, physical strength etc a mixed economy would seem to be a good idea. Then you can probably avoid subsidising wages and create less landfill and C02.


    Are you in need of Urgent Loan Here no collateral required all problem regarding Loan is solve between a short period of time with a low interest rate of 2% and duration more than 20 years what are you waiting for apply now and solve your problem or start a business with Loan paying of various bills I think you have come to the right place I'll advise you can contact us via email at
    Mr Abdullah

  6. I would like to point to something off-topic, but crucial for me. "Sticky wages" assume a societal consensus on wage above starvation level (from my point of view). If the government interferes with a wage floor and subsidise "private firms" the taxpayer has to pay for that bureaucratic overhead (+subsidy). If you want a wage floor and "balanced budgets" who are you going to tax and how efficient would that be?

    Do the "balanced budgetiers", "starve the beasts" and "stiffling bureaucracy" really want decreasing living standards and with that following immiseration? I can only see hypocrisy.

  7. Where did Pieria go?


Post a Comment

Popular posts from this blog

WASPI Campaign's legal action is morally wrong

The foolish Samaritan