Unemployment is a Government Policy Choice

A common misconception is that if everybody was prepared to take awful enough jobs, unemployment would be eradicated automatically, at least eventually, irrespective of the government’s fiscal stance. Embedded in this argument is a misconception that unemployment, overall, can be eliminated through lower wages or deteriorating working conditions. In a capitalist monetary economy, this is not true. To think otherwise is to succumb to a fallacy of composition.

Neoclassical economists made this claim prior to the contributions of Keynes and Kalecki, but it was shown to be unfounded in the capital debates as well as in later work by neoclassical general equilibrium theorists themselves.

Even intuitively there is little reason to expect that an inverse relationship between wages and aggregate employment would hold. A reduction in the price of anything always means two things simultaneously. It means: (i) somebody has to pay less for something they want; and (ii) somebody else is receiving less for providing that thing. At the aggregate level, it means: (i) all of us, taken as a whole, are paying less for the stuff we want; and (ii) all of us, taken as a whole, are receiving less for providing the same stuff. Why would this have any systematic effect on how much stuff will be produced in the economy? It doesn’t, as has been demonstrated formally in the capital debates and later work.

Unemployment is a government policy choice. It occurs when the government fails to maintain demand at a level sufficient to sustain full employment.

Merely redistributing existing income from workers to capitalists by lowering wages has no systematic effect on aggregate demand and employment. The lower wages mean workers are cheaper for firms to hire, but demand for consumption goods may be reduced. So is there more or less impetus for firms to undertake production of consumption goods? Is there more or less impetus for firms to undertake production of investment goods that will increase the capacity to produce consumption goods in the future? The answers to these questions are indeterminate. A mere redistribution of income (from wages to profits or vice versa) has no systematic effect on output and employment.

The bottom line is there are not enough jobs. One person may be able to get an existing job ahead of somebody else by offering to work for less pay or under worse conditions, but this does not alter the aggregate level of employment and unemployment in a predictable way. It may cause employment to increase. Then again, it might cause it to decrease. For instance, redistributing income to capitalists might simply encourage saving and reduce the level of private spending. If so, there will be a multiplied contraction in output and employment.

Unemployment occurs when all of us in aggregate (the non-government) try to save more and spend less than is consistent with full employment given the government’s fiscal settings. The government is uniquely positioned to solve the problem because its spending injects extra demand and financial assets into the economy. Its spending adds financial assets in the form of additional bank balances. These have no offset within the non-government. They are the government’s liability, which a currency-issuing government can cover without difficulty (see here and here). The enhanced financial wealth gives the rest of us greater capacity to spend and save; more spending power. Net saving desires can then be realized alongside higher output and employment. When government chooses not to do this, there is unemployment. That is why it can be said to be a government policy choice.

We as the non-government, in contrast, cannot orchestrate an increase in overall spending power on our own. Private loans are always offset by private debt. Every private transaction creates an asset and offsetting liability. A spontaneous redistribution of existing spending power (for example, a move to higher or lower wages) might – by sheer fluke – reduce the net saving desire and boost demand and employment. But there is no inherent tendency for this to occur.

By injecting additional spending power into the system, the government can render our spending and saving behavior consistent with higher output. That would eliminate the greater part of mass unemployment, and can happen the moment we put our collective foot down and insist on it.

Two qualifications, however, should be noted. First, although it is always possible for a currency-issuing government to enable our spending and saving behavior alongside demand sufficient to sustain full employment, this does not mean that this action will always be appropriate. An excessive saving desire might reflect extreme inequalities of income and wealth. The wealthy typically spend a lower proportion of their income than middle and low-income individuals. By moderating distribution, government can encourage stronger private spending and lower saving for a given fiscal stance. It will then take a smaller government deficit to sustain demand consistent with full employment.

Second, as the economy gets very close to *true* full employment (meaning 2% unemployment to allow for frictional factors), inflationary pressures are likely to emerge. To achieve true full employment alongside low and stable inflation, a job-guarantee program would need to be integrated into the institutional structure of the economy to act as an automatic stabilizer and nominal price anchor. In the absence of a job guarantee, price stability can only be achieved through the deliberate maintenance of some unemployment. Governments of the neoliberal period have tended to take that latter choice.

Further Reading

For more on the notion that government can net spend in a manner compatible with our spending and saving behavior, see:

Parable of a Monetary Economy

A discussion of the job-guarantee proposal and its superiority to the neoliberal NAIRU approach is contained in this post:

Short Note on the Policy Target

Regular readers will be aware that my own preference is for a combined ‘job or income guarantee‘ in which individuals can opt for a job or be satisfied with a basic income. Whether there is a basic income or not, a job-guarantee component is necessary if we want true full employment alongside price stability. With a basic income but no job guarantee, we could have strong employment, price stability and income security for all, but not true full employment. Including a job guarantee would enable simultaneous achievement of all these goals.


11 thoughts on “Unemployment is a Government Policy Choice

  1. Also, new look is great. Need to move heteconomist label off your face, Peter. In other words, stop being stupid.

  2. Trouble is, I can’t see how to make sure the heteconomist label shows up at all unless I leave it there. Then we’d forget where we were and escape through a wormhole, leaving TPTB totally thrown. I’m working on it, though. It’s a very new theme, and my tech capabilities are near zero. Besides, look at Warren. He has a whole post in front of him (you may need to wait a second or two for him to appear):

    Warren Behind An Entire Post

  3. Hello peterc and others,

    It’s been a while (several years) since I started studying MMT and it’s been quite a while of following this blog. However, up until now I still can’t get how unemployment is a government policy choice, if we ignore JG. I mean, what NFA has to do with all of this?

    Generally, could you please respond to Cullen’s piece that Trixie linked to?

    I am rather perplexed by all this so that I can’t even formulate a specific question, but it just doesn’t add up in my head how no net savings would lead to no unemployment. I understand how it would lead to less unemployment, but none? Hasn’t the U.S. and other countries had negative net savings at some points in their history? I hope you get what I mean!

  4. “We as the non-government, in contrast, cannot orchestrate an increase in overall spending power on our own. Private loans are always offset by private debt.”

    I don’t see how that makes sense. Say a bank extends credit by making a loan and creating a new deposit. The deposit is an asset to the borrower which is matched by the borrower’s debt. But the deposit also represents new ‘money’ which the borrower can spend that didn’t exist before. This is an increase in overall spending power, isn’t it?

  5. Yes, I’ve seen that from Warren. How do you people dress yourselves?

    And no, don’t take the pic away, it looked good. Jesus. Simply alter the image with ‘heteconomist’ further up/down on the pic, and ditch the label altogether.

  6. Philippe and Edgaras. Thanks for joining the discussion. I started to respond to both of you and, indirectly, Cullen and it became too long for a comment, so I have put it up as a new post.

    Thanks, also, Trixie, for notifying me of Cullen’s post. I’ll get to the formatting in time.

  7. I think you’re ignoring the fact that labour has substitutes. At a high wage rate, some employers might prefer to mechanise certain tasks, or forego them entirely. I accept that a lower total wage bill across the economy will mean proportionally reduced demand for goods and services, and thus be self-defeating, but higher levels of employment can be achieved by offering new jobs that warrant lower wages.

    You mentioned in another article how government could simply guarantee employment at minimum wage, funded by fiat currency, and that would put paid to unemployment. But why do it at minimum wage? We could all be millionaires, if government guaranteed employment at rather more than minimum wage, funded by fiat currency.

  8. Michael G: The argument doesn’t ignore that machines and workers can be substitutable. It simply takes into account that the capital debates demonstrated this does not ensure a tendency toward the full utilization of the neoclassical ‘factors’ of production.

    Cambridge Capital Controversy
    Capital debates: A brief introduction
    Lucas in context, Keynes out of context
    Vernengo on Keynes vs Neoclassical Synthesis
    More on Keynes vs the Neoclassical Synthesis

    The job guarantee (JG) is intended to function as a nominal price anchor and automatic stabilizer. The JG wage sets the floor for the broader economy. Private-sector and other public-sector employers can entice workers out of the JG program by offering more attractive jobs. During an upturn in the broader economy, the JG provider would not compete on wages, so JG workers would be drawn into private and other public employment. In a downturn, the JG puts a floor under demand and helps to counteract any deflationary pressures.

    So, one reason the wage is not set at infinity is that the value of the currency is maintained by delivering low and stable inflation. The resource constraint can also be viewed as an inflation constraint. It is necessary for the government’s fiscal policy to be kept consistent with non-government spending and saving behavior, as explained in the post, but the JG provides an additional anchor to the system.

    Value of the Currency
    A Short Note on Currency Value

  9. In reference to the job guarantee, does not the existing unemployment benefit or work for the dole programme act as a form of job guarantee?

  10. The job guarantee, as proposed by MMTers, is to provide a full-time job on demand at minimum wage and standard benefits. A part-time position could be chosen if that was what the person demanding the job preferred.

    MMT proponents of a JG distinguish the program from workfare such as work for the dole, but clearly there is a political danger of such programs being reduced to that in implementation.

Comments are closed.