Winner of the New Statesman SPERI Prize in Political Economy 2016

Saturday 22 April 2017

Breaking the ‘strong economy’ narrative

My last post talked about the gap between the macroeconomic narrative in the UK media (‘mediamacro’) and macroeconomic facts. The gap is created or encouraged to a considerable extent by narratives employed by the political right. So how might that change, to let reality back in?

As with other things, Labour under Miliband had the right idea but did not follow it through. They talked about a ‘cost of living’ crisis, but in doing so they implicitly suggested this was some unfortunate by-product of a strong economy. The aim should be to redefine a strong economy as one that delivers solid real wage growth.

To do so makes perfect sense in current circumstances, when we have just had a policy-induced large depreciation in sterling. GDP measures the output produced in the economy, but not how much people in that economy can buy. Welfare depends on the latter, not the former.

It also makes sense if real wages have fallen because workers have priced themselves into jobs, by in effect discouraging firms to invest in labour saving machinery. Boasts that employment is at record levels make no sense in that situation, because high employment comes from lower wages rather than from additional output. [1]

I have stressed in the past (including my last post) how weak recent UK performance has been by historical standards. But a favourite trick of the government is to make international comparisons, of GDP rather than the more appropriate GDP per head. So how does our economy look if we focus, more appropriately as I argue above, on international comparisons of real wage growth?

Luckily the ILO and Geoff Tily have already done the spade work. Here is a chart for all countries, with blue denoting OECD countries.

International comparison of average real wage growth since the crisis 

Source: Geoff Tily, ILO. 

Among OECD countries the answer is striking: only Greece has seen real wages falls greater than the UK. The UK is second best among the OECD at achieving a decline in real wages! Geoff looks at data from 2008, but a quick check suggests the result holds good if we start in 2010 instead.

The data in this comparison only goes to 2015. You could, rightly, argue that 2016 was a better year for the UK, but then you would have to address what will happen to real wages this year and next. [2] You could argue that this poor performance was a consequence of the 2008 depreciation (which had lagged effects): again you would be right, but the Brexit depreciation which is not yet in these figures is just as large.

Either way this data provides strong evidence of just how terrible UK economic performance has been over the last several years. [3] What is more, unlike GDP, it is data that directly relates to the experience of ordinary people. But as Miliband found out, to quote this data is not enough. What you need to do is start proclaiming that the UK economy under a Conservative Chancellor has performed worse than any other OECD economy besides Greece. Just that, no caveats, no qualifications, no ‘cost of living’ label. Only that way will you begin to shift the narrative that we have a strong economy.

[1] If you are worried that this might help justify calls to reduce immigration, fear not. What they show is that policymakers failed to create an adequate level of aggregate demand: another consequence of austerity.

[2] If we look at the ONS series for real average earnings, normalised to 100 for 2015, it was at 101.8 in May 2010, and in February 2015 it is 100.3, a fall of 1.5%

[3] It has even been fact checked: see here.              


  1. Asked how he wanted to be remembered as a politician, George Osborne last week said: "As someone who left the country in a better state than he found it."

    Of course the interviewer - John Pienaar I think - didn't call him on it.

    Has there ever been a more mendacious clown in British public life?

  2. Informative data and framing. Thank you. I agree, this should be trumpeted load and clear with no frills.

  3. I pointed out in your last post that GDP per capita is a poor metric unless you consider demographic changes. You were cherry picking statistics to make a cheap political point. Your analysis in this post is more rubbish.
    Since May 2010, the UK has added about 2.5 million jobs of with about 2 million foreign born in employment. These jobs are disproportionately low paid, so the arithmetic alone will have a significant affect on average wages. People already in employment will have achieved somewhat higher wage growth and many more young people and unemployed will have achieved higher welfare. Many of the OECD countries have higher wage growth at the cost of higher unemployment. Most of the foreign born workers will have been earning much lower wages in their own countries. There is a good argument for either excluding foreign workers from this kind of analysis or somehow including their original earning levels in the overall measurements.

    1. The value of different variables depend on what are looking to do. There is no sense in which GDP per capita is a poor metric without further qualifications.

      If the insistence is on profiting to many people, then GDP per capita is problematic because means are very sensitive to outliers and income distributions have many of them. However, if we just look at median wage growth or median income growth as suggested by another post, we also end up ignoring the high end of the income distribution.

      Usually, we use many different variables in conjunction with sometimes rather sophisticated statistical tools to answer questions related to policy evaluation because answers are hard to get.

  4. «redefine a strong economy as one that delivers solid real wage growth.»

    That aim is interesting, especially it is about unit wage (per hour) growth, rather than total wage (per year) which can grow by simply increasing working hours.
    The problem with that redefinition is that the Economics profession have defined for several decades wage growth as "inflation" and asserted that the main purpose of Economics policy is avoiding wage growth, and a strong economy has been defined instead as one with robust stocks and property price growth, thus improving aggregate "productivity" by reallocating resources from "low productivity" wage workers to "high productivity" rentier owners.

    It is good that our blogger is now advocating the notion that wage growth and improving living standards for most of the people is the main measure of economic policy success, but there is a mountain of contrary attitude to overcome.

    «But a favourite trick of the government is to make international comparisons, of GDP rather than the more appropriate GDP per head»

    That is a favourite trick of Economists in general, as aggregate GDP growth is positive for asset price growth. As to GDP per head growth it is not quite the same as wage growth, as GDP per head growth can go mostly to property incomes or to very high wage incomes. In general for wage growth it is better to look at *median* wage (or even income, if not available) growth in the 25-64 or 25-55 age range.

    Anyhow apart from the bar graph above, I find the following graph of GNI (rather than GDP, even if for most countries they are similar) per head at PPP constant 2011 dollars for some groups of european countries quite interesting to look at:

  5. As to median wages (and that should be per hour ideally), I have briefly searched EuroStat for that, but not found yet. I found instead the famous series for the USA, 1979-2016, of weekly median wage deflated by the (ever optimistic) CPI, which has been oscillating around $320 per week 1979-1997 and around $340 per week 1997-2016, and it would be interesting to see the same graph for the UK:

  6. I don't know how to fix that problem, seriously. Most journalists went to college and they're still apparently incapable of picking up on how politicians purposefully frame problems. Or, else, they see it but choose to not spill the beans -- or, maybe, they are told not to spill the beans. It wouldn't be the first time we hear of people being censored in a very wicked manner. Regardless of the reasons, we are unlikely to see them ask real questions to politicians in the near future.

    What you present on this blog, most of the time, seems very straightforward to me. But do people really approach the matter in a way that allow them to be convinced? My experience of people outside the academia, and sometimes also within, is that they approach discussions as debates and talk to win, not to understand. How do you get someone in that mind set to turn their head and look at what's going on is beyond me.

  7. Ah found almost what I wanted, average labor cost per hour, snapshot of graph for several european countries

    Since it is *average* labor costs it is probably quite different from median.

  8. I fully agree with the previous comment that GDP per head is misleading when you add 2.5 million jobs to the economy. You should the stats without these jobs or count them with a income of zero for "before". Also, I think economy will take a back seat in this election because (1) brexit showed us that "identity politics" is back (as the excellent "Road to Somewhere" explains) and (2) the Labour party was taken over by psychopath marxists with the intellectual level of an ex-polytech dropout. Even if they had all the right policies, nobody sensible can vote for them

    1. There is a presumption that migrants worked for below average wages. Do you have any evidence to back that up?

    2. Search for "Foreign nationals make up 17 per cent of Britain's 2.45 million child tax credit claimants, figures from HMRC show". Tax credit claimants have below average wages, and working migrants are over-represented on tax credit claims.

    3. Here is some data on average wages:

    4. It covers 1993 onwards and says:
      "Wages of migrants have been converging with those of the UK-born, but A8 migrants have lower wages"
      And River was specifically talking about the jobs added since the financial crisis.

      For the record I'm in favour of free movement for work, and would like to see market controls brought in, not controls set by the government.

  9. Simon -- Are we not just seeing the great financial bubble working in reverse here -- ie the bubble artificially boosted GDP in the UK during the boom, and has had the opposite effect since 2008. Because the UK has a larger financial sector than other economies, it has lost relatively more in the bust.

    1. There is very little evidence of a boom in 2007: look at consumption, or inflation.

  10. You put the issue in terms of political perception but I think the failure is actually more basic than that.

    We expect continuous growth and yet, as people like Bob Gordon have pointed out, these things come in cycles and are not continuous by nature.

    Growth is composed of two main elements:population and productivity. Growth in the working age population in the developed economies is falling and this effects not only the host country but those who they trade with.

    Productivity has languished, perhaps explained by the Bob Gordon thesis.

    So what do we expect? The continuation of BAU!

    It seems to me the failure is far more than being disingenuous with the GDP figures; it's a far more fundamental failure to explain the economic facts of life and where reasonable expectations should rest. How much discussion is there about the secular increase in the cost of energy (been going up for years) and the implications of robotics/AI - virtually none, and yet these will be huge influences in our economic future within the next fifteen years or so.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.