Monday, 21 July 2014

The clash of micro and macro

As I said in a recent blogpost, failing to provide microfoundations for a macroeconomic argument doesn't make the macroeconomics wrong. Conversely, providing lots of lovely microeconomic detail - right down to the "I met a man" level - does not necessarily add up to a convincing macroeconomic argument. 

So here is Chris Dillow taking Tim Montgomerie to task for claiming that the UK's remarkable employment performance is due to the Coalition government's welfare reforms. Montgomerie isn't the only one making this claim: Fraser Nelson does so too in more detail, in an op-ed in the Telegraph. Both Tim and Fraser say that the Coalition's welfare reforms, by forcing lots more people into work, have somehow created a massive jobs boom. Say's Law, applied to labour markets? Hmm. Maybe I'm a bit fonder of microfoundations than I thought. I want to know where these workers really come from - the increase looks too much to be entirely due to benefit reforms. And I want a proper explanation of the jobs boom. Simply saying "there are more workers, therefore there are more jobs" doesn't do it for me. 

Here is Giles Wilkes providing a pretty good explanation for the increase in workers. No, it isn't just benefit reforms. A quarter of a million or so pensioners have returned to the labour market, for starters - and no-one is suggesting that benefit cuts are forcing them back to work (though the Bank of England might be).  Tax changes have improved incentives to work at the margin, which would draw some people into the workforce. Tax credit changes have forced some people to work longer hours in order to qualify. And above all, the squeeze on living standards caused by falling real incomes in recent years, on top of a horrible shock to both incomes and wealth, has forced people who weren't working to do so and part-timers to increase their hours. 

But what about that increase in jobs? Well, Giles says that as the supply of labour increases, we would expect its price to fall, encouraging employers to create more jobs at lower wages. And as Chris points out, the effect of pushing people into work by making life on benefits either impossible or totally miserable is to force down wages. I know I have said this many times already, but it's worth repeating it. Workfare depresses wages not just for those being forced to earn their benefits, but for everyone. So it may be that making life on benefits more miserable does increase jobs, but at the cost of lower wages. And the re-entry to the workforce of people who have a basic income (pensioners, married women with children) is also likely to depress wages, since these people don't usually face the benefit withdrawal trap that forces people to choose between benefits and employment, so can afford to take lower-paid work.

So to paraphrase Chris's conclusion: if welfare reform significantly increases the supply of labour, and all that labour can be productively employed, then in the absence of strong economic growth, welfare reform depresses wages - which might not be the message the Conservatives really want to give. They would do better to point to improving economic conditions as the principal driver of rising employment. Despite Fraser's claim that Coalition welfare reforms are popular, the macro argument seems more likely to win votes than the micro policies.

But Simon Cooke nevertheless criticises Chris Dillow for ignoring the human stories underlying rising employment. To him, the macro argument that improving economic conditions and/or falling wages are the primary cause of rising employment implies that micro policies aimed at helping individuals into work are a waste of time and money. And he disputes this, using as examples some of the real human stories that he has encountered in his work as a local politician. I don't think many people would disagree with his conclusion: getting the long-term unemployed into work is indeed hard, not least because of the barriers that society - often with the best of intentions, such as protecting the public from possibly dangerous ex-cons - puts in their way. We should respect those who devote time and energy to helping these people into work.

But Simon's argument that there is a fundamental disconnect between macro and micro policies does not follow. Chris's macro argument actually supports Simon's micro policies: if economic conditions improve and/or wage levels fall, then helping the most disadvantaged people into work becomes easier. The Work Programme seems to be doing a decent job, though it is much less clear that Help to Work justifies its cost. But it has been greatly hampered by the prolonged economic downturn. As the economy improves, the Work Programme's results will improve too - though admittedly, some of the people it will "help into work" would have found work anyway. Sometimes improving economic conditions really is all that is needed.

The Coalition has rightly been criticised for trying to force people to work who were not able to do so, for completely fouling up the administration of the fitness to work tests, and for some frankly pointless and even harmful policies (the "bedroom tax" springs to mind). But the biggest problem is that they tried to do all of this at a time when economic conditions were utterly unhelpful. If Giles is right, then the changes intended to "make work pay" (tax changes as well as benefit reforms) have actually pushed down wages and made life harder not just for the thousands who were forced into work, but for millions. That is quite an indictment of a government.


  1. One bit of evidence that supports Montgomerie is that Bill Clinton slashed welfare and a significant proportion of the allegedly unemployed soon found jobs. But obviously that bit of evidence doesn’t prove Montgomerie’s case.

    “Workfare depresses wages not just for those being forced to earn their benefits, but for everyone. So it may be that making life on benefits more miserable does increase jobs, but at the cost of lower wages.” I suggest it’s only wages at the lower end of the scale that get depressed by workfare: i.e. workfare employees are not exactly competition for qualified plumbers, bricklayers, lawyers, doctors, etc.

    1. You would think so, but in fact it depresses wages far out along the curve. I think that is because we generally see workfare in slack labour markets where there is unemployment among many skilled categories as well as the unskilled. Where there is workfare, the skilled unemployed face being forced into unskilled (demeaning) jobs, so will accept skilled jobs at lower wages. Of course, there are always some skill categories that don't suffer this effect, and their wages may continue to rise even as others fall. The labour market is not homogenous.

  2. On second thoughts, there is a reason for thinking that workfare does no even necessarily reduce wages for the less skilled. In fact it might INCREASE them. The reason is this.

    The marginal product of labour declines as employment rises (i.e. as unemployment falls). That’s because as unemployment falls, the SUITABILITY of the unemployed for vacancies declines. That unsuitability at some stage prevents a further reduction in unemployment (i.e. the economy hits capacity or NAIRU). However, if workfare makes up for that unsuitability, then TOTAL NUMBERS EMPLOYED will rise (i.e. NAIRU declines). Ergo GDP rises. Ergo potentially EVERYONE’S wage rises.

    1. No, there is a logical disconnect there, Ralph. Forcing down wages may increase employment, and hence GDP, but that does not necessarily follow through into higher wages. Just look at Germany.

      Furthermore, forcing the less productive into work has a negative effect on productivity. The UK's puzzle is rising employment, falling real wages AND poor productivity. If the average marginal product is falling, then wages will not rise even if GDP is rising: higher employment means the TOTAL wage bill rises, mopping up the GDP gains, without INDIVIDUAL wages rising at all. It's the macro/micro clash again. At the macro level, total wages rise. At the individual level, they do not - indeed they may even fall. And as labour becomes cheaper the incentive for employers to do anything about low individual productivity, for example by capital investment, declines. What you end up with is a low-skill, low-productivity, high-employment economy. Is that what we really want?

    2. “Forcing down wages may increase employment, and hence GDP, but that does not necessarily follow through into higher wages. Just look at Germany.”

      I’m puzzled as to exactly how it’s possible to raise GDP while wages fall. The only possible explanation (assuming all else equal – terms of trade in particular) is that profits rise AT THE EXPENSE of wages. But I don’t see that introducing workfare has a big effect on profits: a significant proportion of employers regard workfare employees as being more trouble than they’re worth.

      Moreover, even if introducing workfare DOES increase profits at the expense of wages, market forces ought ultimately to return profits to their “normal” level, to use economics jargon.

      “forcing the less productive into work has a negative effect on productivity.” Agreed. Workfare employees are relatively unproductive, so average productivity declines when workfare is introduced, but that can be consistent with rising GDP. To illustrate, if numbers employed rises by 1%, while the output of those extra employees is half the average, then average productivity declines, while GDP rises.

      “If the average marginal product is falling, then wages will not rise even if GDP is rising: higher employment means the TOTAL wage bill rises, mopping up the GDP gains…”

      Again, I’m baffled as to how it’s possible for wages to fall when GDP is rising. Where does the extra output go? It can only go to profits – there’s nowhere else for it to go. And for reasons I gave above, I don’t see that there’s much effect on profits.

      “At the macro level, total wages rise. At the individual level, they do not - indeed they may even fall.” I’m baffled again. How is it possible for “total wages” to rise (I assume you mean in real terms) while “at the individual level” wages fall? Total wages are simply the sum of “individual wages” aren’t they?

    3. Ralph. Let me give you a simple example.

      Supposing I am a widget manufacturer who employs 20 people at £10 per hour. Instead of investing in capital equipment to improve the productivity of my existing staff, I take on 10 new staff at a lower rate of £8 per hour. My wage bill has risen, clearly. My turnover has increased. My profits MAY have increased, though my failure to invest might mean I only just barely cover my costs and my profits remain low. Result is is higher employment, higher GDP, flat wages (or falling, actually, since I recruited my new staff at a lower rate) and flat profits. Rising GDP and rising employment do not necessarily translate directly into either higher profits or higher wages.

    4. Frances,

      In your widget example, I think you’ve ignored the question as to who was supporting the 10 new employees before they were taken on. When the new employees are taken on, EXISTING employees pay less tax to support the previously unemployed. So everyone’s after tax wage / income will rise. Or at least AVERAGE incomes will rise.

      Even if there’s no formal social security system, the unemployed INEVITABLY live off the employed by begging, borrowing or stealing as they did 100/200 years ago.

      On a different point and re the title of your article (Clash of Macro and Micro) that is apt because in working out the effects of workfare one has to think of both macro and micro effects. Richard Layard (labour market economist if ever there was one) once referred to the “quagmire” that lies between macro and micro.

    5. What makes you think that the new staff are a) unemployed b) claiming benefits? I did not say that. They might simply be inactive people returning to the labour force - we've had a lot of those.

      Also, if you recall we have something called Tax Credits, which are paid out of general taxation. If these people claim Tax Credits to top up their wages, who do you suppose pays for that?

      There is actually no direct pass-through from benefit reduction to tax cuts. The workers in my widget factory aren't going to get tax cuts because I have taken on 10 new staff. Even if every factory in the country took on 10 new staff, they still might not get tax cuts. Government might simply find something else to do with the money.