Winner of the New Statesman SPERI Prize in Political Economy 2016


Wednesday, 30 May 2018

Nominal wages are not real wages, and why it matters in the UK



This post is just an extension of a recent tweet from Chris Dillow. I think it is worth writing more about it because it reflects on an issue that is widely misunderstood, by some on both the left and the right. Here is the share of employees compensation and corporate profits in total income since 1948. Note that everything below is about the UK experience: the US is different.

I have gone back to 1948 to show that the wage share can change, and has fallen from the 1950s until the end of the century, from 60% to 50%. But that has not been accompanied by a rise in the profit share, which has stayed pretty close to 20%. The missing pieces, that are the counterpart to the fall in the labour share over this period, are income going to indirect taxes, self employment and unincorporated businesses.

The key point I want to make is that neither the wage share or the profits share has changed over the last 15 years. This busts two myths that you will often see.

Myth 1 Immigration has kept real wages low.

There was a large increase in non-EU immigration at the end of the 1990s: if anything the wage share increased at the same time. The second large increase in immigration, this time from the EU, was from 2004, and there was no noticeable impact on the labour share. Immigration may have been depressing nominal wages, but those lower nominal wages were allowing lower output prices, leaving real wages unchanged. This is consistent with the econometric evidence that immigration has no significant impact on real wages.

This evidence is often dismissed in two ways. The first is that it does not correspond to workers 'lived experience'. But that experience reflects either the impact of immigration on their own nominal wage, or falls in real wages which reflect the lack of productivity growth and sterling’s depreciation. It is just possible that immigration might have discouraged firms from investing in higher productivity techniques, but it is rather more likely that immigration has allowed firms to produce in the UK who would otherwise have produced abroad. For other reasons why intuition on immigration may be misleading see here.

The second way that some people argue that there ‘must be’ a link between real wages and immigration is to invoke simple supply and demand. That is just fallacious: immigration shifts the labour supply curve but it also shifts the demand curve. A slightly more sophisticated argument is that the demand curve does not fully shift to compensate for greater supply immediately after an increase in immigration because it takes time to invest, but if that was the case we would see a temporary fall in real wages and a rise in the profit share following periods of immigration, and we do not.

Myth 2: Real wages have fallen because labour is now weak compared to employers.

Since the Global Finance Crisis (GFC) and subsequent recession, nominal wage growth may be slow because the labour market is weak, but the data shows that employers are not taking advantage of this to increase the profit share. Low wages are being passed on into lower prices.

There are two main reasons why real wages are currently low: almost non-existent productivity growth since the crisis and the depreciation in sterling which has raised the cost of imports and therefore consumer prices.

Note that I am not at all saying that the labour market is not as weak as it appears. In many areas conditions of employment seem to have deteriorated since the GFC. What I am saying is that real wages depend on prices as well as nominal wages, and in the UK at least there seems to be sufficient pressure on firms to pass on low wages into low prices, leaving the relationship between real wages and productivity unchanged.

It is also important to point out that the wage share is different from median wages. As the study by Pessoa and Van Reenen I examine here shows, median wages from 1972 to 2010  declined relative to the average compensation because of rising non-wage benefits and rising inequality. A good part of this rising inequality reflects incomes of the top few percent.  

It is often difficult to convince those on the left that weakening labour power over wages can be a good thing, if lower nominal wages are passed on to lower prices. It can be a good thing because it allows the central bank to raise demand and therefore output by more than they otherwise could while keeping inflation stable. It reduces the sustainable unemployment rate: the NAIRU. Furthermore lower wages are more likely to be fully passed on into lower prices if the goods market is highly competitive, and that is more likely to happen if the economy is open to overseas trade.


12 comments:

  1. Interesting as always, Simon, but I'm not at all sure what can be concluded from essentially static profit and wage shares over the past fifteen years. These are shares of whole-economy GDP which is a pie that has grown over the period, but the population, the number of workers and the number of employees (these latter being the recipients of the wage share) have grown rather more. So the wage share slice has to be divided up among more people: each employee is receiving a smaller piece of GDP than they would have previously. On the other hand, the profit share does not necessarily have to divided up among more recipients of profit. Indeed, although the profit share has been stable, *profitability* certainly in the services sector has been noticeably higher since 2013 than in the preceding decade. So a relatively steady relationship between wage and profit shares in GDP does not imply that returns to labour have held up at the individual level. You do allude to this later on by pointing out that the wage share is not the same as median wages, but it is seems to me that this (and other distributional aspects) merits rather more than an aside if the 'myths' are to be busted.

    ReplyDelete
  2. The demise of phony experts: macroeconomics is provably false
    Comment on Simon Wren-Lewis on ‘Nominal wages are not real wages, and why it matters in the UK’

    Simon Wren-Lewis discusses the relationship between UK wage- and profit share, real wage, employment, and the NAIRU in order to bust two myths about immigration and employment. This is a futile exercise because macroeconomics is false since Keynes. After-Keynesians have not realized it until this day.#1

    Roughly speaking, economists never got their foundational concepts straight. Here is the evidence from the General Theory: “Income = value of output = consumption + investment. Saving = income − consumption. Therefore saving = investment.” (p. 63)

    This two-liner is conceptually and logically defective because Keynes never came to grips with profit: “His Collected Writings show that he wrestled to solve the Profit Puzzle up till the semi-final versions of his GT but in the end he gave up and discarded the draft chapter dealing with it.” (Tómasson et al.)

    Because the foundational concept profit is false, the whole analytical superstructure is false. Since Keynes, macroeconomics is scientifically worthless.

    Macroeconomic profit Qm is not a flow like wage income Yw but the difference of flows, i.e. Qm≡C−Yw in the most elementary case. It is methodologically not admissible to add wage income Yw and profit Qm together and to call the sum total income. This is the Humpty Dumpty Fallacy.#2

    Because the Profit Theory is false, Distribution Theory is false and the concept of profit “share” is nonsensical.

    From the correct macroeconomic axioms follows the Profit Law as Qm=Yd+(I−Sm)+(G−T)+(X−M). Total income is given as the sum of wage income Yw and distributed profit Yd.

    Employment Theory is false by logical implication. The most elementary version of the correct macroeconomic Employment Law is shown on Wikimedia.#3

    From this equation follows, inter alia, that overall employment INCREASES if the average wage rate W INCREASES relative to average price P and productivity R. This is the opposite of what standard economics teaches.

    As Simon Wren-Lewis argues: “It is often difficult to convince those on the left that weakening labour power over wages can be a good thing, if lower nominal wages are passed on to lower prices. It can be a good thing because it allows the central bank to raise demand and therefore output by more than they otherwise could while keeping inflation stable. It reduces the sustainable unemployment rate: the NAIRU.”

    It is difficult to convince microfounded economists, who are hopelessly trapped in the Fallacy of Composition, that the average wage rate must rise relative to price and productivity in order to reduce unemployment and that NAIRU has always been a NONENTITY.#4, #5

    Right policy depends on true theory: “In order to tell the politicians and practitioners something about causes and best means, the economist needs the true theory or else he has not much more to offer than educated common sense or his personal opinion.” (Stigum)

    Economists do NOT have the true theory. Macroeconomics is axiomatically false. Economists do not know how the price and profit mechanism works. Economic policy guidance has no sound scientific foundations since Adam Smith/Karl Marx. There never was such a thing as an economic expert.

    Egmont Kakarot-Handtke

    #1 Why Post Keynesianism Is Not Yet a Science
    https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1966438

    #2 For details of the big picture see cross-references Profit
    http://axecorg.blogspot.com/2015/03/profit-cross-references.html

    #3 Wikimedia, Employment Law
    https://commons.wikimedia.org/wiki/File:AXEC62.png

    #4 Keynes’ Employment Function and the Gratuitous Phillips Curve Disaster
    https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2130421

    #5 For details of the big picture see cross-references Employment
    http://axecorg.blogspot.com/2015/08/employmentphillips-curve-cross.html

    ReplyDelete
  3. I am a believer in the evidence that says that immigrants have not reduced wages (and support immigration more broadly, for philosophical reasons as well as economic). But surely a constant wage-share of income is consistent with immigration having a negative effect on wages if that wage-share is split between more people? In fact, isn't that exactly what you would expect from a Cobb-Douglas model: labour share would be constant regardless of amount of labour used (marginal product would reduce as more labour added).

    ReplyDelete
  4. I think this needs to be taken forward with a distributional analysis, to see how different income groups have fared. Has this been done?

    ReplyDelete
  5. “Immigration may have been depressing nominal wages, but those lower nominal wages were allowing lower output prices, leaving real wages unchanged.”

    We have evidence of this in the forecourts of former garages where automatic car wash facilities lie idle whilst immigrant workers offer manual car washing. You might argue that this is a good thing but many would disagree. The issue here is that the beneficiaries of these lower prices are those who can afford to pay for regular car washes, (if they can afford a car). Furthermore if the workers are claiming tax credits etc then non-car owners are cross-subsidising car owners. No doubt too the surfeit of foreign waiters and baristas has benefitted those who can afford to regularly dine out.

    “This evidence is often dismissed in two ways. The first is that it does not correspond to workers 'lived experience'.

    Comparing average nominal wages with average prices can only ever tell as part of the story. We cannot assume that all workers are equally impacted by wage depression nor that they can all equally benefit from reduced prices. The issue is whether immigration has had a disproportionate adverse effect on the poorer members of our society and thereby raised inequality. The lived experience is relevant here.

    “Immigration may have been depressing nominal wages” or of course it may have prevented the salaries at the lower end of the scale from rising at a rate much faster than inflation in response to a shortage of labour.

    “Immigration shifts the labour supply curve but it also shifts the demand curve.”

    This counters the arguments being put forward by Remainers. They argue that immigrants are needed to fulfil employers’ needs for workers. But here you are suggesting that employing more immigrants will not address the shortfall between the demand for and supply of labour, since the supply of new labour will create a further proportionate demand.

    Let’s be clear Tesco is in favour of immigration because a 1% in increase in population represents a 1% increase in its potential customer base, whereas Tesco customers would like to see is a 1% increase in Tesco turnover due to a 1% increase in their own prosperity . A 1% increase in population would inflict on Tesco customers a 1% increase in traffic on the motorways, a 1% increase in overcrowding on the railways and a 1% increase in the competition for housing.

    ReplyDelete
    Replies
    1. "We cannot assume that all workers ... can all equally benefit from reduced prices."

      Having read the comment by Blissex, I realise that there is prime example of this.

      Let us suppose that for residential care homes lower nominal wages are passed on as lower fees. Since the occupants of residential homes are predominantly not working for a living, lower residential fees will not benefit people who are in work.

      Delete
  6. I am not arguing with the conclusion, but am not sure that the evidence proves the conclusion.

    Consider a scenario. There are 35 workers. The Nominal Wage is LS/35. An new worker comes in (the immigrant). The nominal wage is LS/36, but LS has changed in real terms by the addition of addition production. If the additional production were equal to average production, then real wages would remain the same, but economics says that the production increase would be only the marginal production (since the higher value jobs would be staffed first.) Thus, you would expect real wages to go down, and productivity to go down, at least in the short run.

    (Yes, this rather simplistic analysis ignores certain factors. Fixed government costs do not rise because of additional workers. The interest on the national debt does not go up because of immigrants. Likewise Britain need not spend more on defense because some immigrants moved in. Additionally, if the immigrants have skillsets not common to locals (nurses) then the productivity of the immigrants may be higher than the locals, raising national income more than the marginal productivity assumed above.Likewise, if the immigrants use less services then that affects the calculations.)

    But as a generic proof - we increased the number of workers, the proportion of national income going to labor did not fall, therefore the real wages did not fall, is only true if the economy grew more by a percentage greater than the increase in workers.

    ReplyDelete
  7. «Furthermore lower wages are more likely to be fully passed on into lower prices if the goods market is highly competitive, and that is more likely to happen if the economy is open to overseas trade.»

    The whole argument and this conclusion in particular are so ridiculous that I hope our blogger deletes it.

    Consider the situation where "ceteris paribus" (as if!) and the entirety of lower nominal wages became lower nominal prices. This results *obviously* in lower real wages, because while the fall in nominal wages is suffered only by wage-earners, the fall in nominal prices benefits also earners from fixed-income or variable-income assets such as pensioners and shareholders. This is startlingly obvious, and the numbers of beneficiaries from fixed and variable income assets is far from negligible (has our blogger recently retired with a defined-benefit pension? :->).

    Consider symmetrically a fall in nominal asset prices because of a fall in the nominal income associated with them: would that increase the real earnings of asset owners? :-)

    The ridiculous propaganda that lower nominal wages in a country and imports from other countries with lower nominal (after exchange rate conversion) result in the same or higher real wages because of lower prices is just neoliberal drivel as it rather simply results in redistribution from a country's wage earners to its asset owners, and to the workers of exporting countries.

    The less ridiculous defence of lower domestic real wages thanks to immigration and higher imports from low-wage countries is that the welfare loss to domestic workers is much smaller than the welfare gain to low-wage foreign workers who emigrate or produce for exports and that is almost certainly true.

    ReplyDelete
  8. In regards to the benefit of weakened labor power if the "savings" are passed on to consumers as lower prices, people could have a problem with that form of higher real wages for political reasons (in the realpolitik sense rather than ideological). Weaker labor bargaining power means weaker political power for labor as a class. With weaker labor bargaining power the broad benefits to labor are reliant on capital "allowing" them to occur. The "benefits" of real wages are entirely in the hands of capital who could funnel the savings from lower wages to profits. Obviously it's not as easy as that, but not every sector could possibly be a competitive market. And as Marshall Steinbaum and others have shown (admittedly in the US context) capital has been able to increase the areas where they have monopoly and monopsony power by using the increased political power that has come from a decrease in labor bargaining power.

    Relying on lower prices to raise the real wage depends on "trusting" capital to allow competitive markets even though they might have the political power to decrease competitiveness in a way which would help them tremendously. What about the last few decades has shown capital as a class can be trusted to support society as a whole in order to benefit others even when it might hurt capital owners?

    ReplyDelete
  9. A few points:
    - Using wages deflated by CPI does not reflect the impact of housing costs. Hence people are actually a lot poorer
    - 10% is a massive fall in wages share. If labour were to regain 10% of today's GDP, that would be £4k per UK resident
    - Agree that external factors have impacted real wages in the period presented, but in the face of the same external pressures, the profit share has increased, showing just how weak labour is currently vs capital
    - Disagree that real wages are mainly impacted by lack of productivity gains. In the period 1973-2014, productivity gains were 72.2%, while hourly compensation increased by 9%. Since 1979 in particular, the two have decoupled.

    ReplyDelete
  10. "I have gone back to 1948 to show that the wage share can change, and has fallen from the 1950s until the end of the century, from 60% to 50%. But that has not been accompanied by a rise in the profit share, which has stayed pretty close to 20%. The missing pieces, that are the counterpart to the fall in the labour share over this period, are income going to indirect taxes, self employment and unincorporated businesses."

    One or two of us - prompted by the very same Chris Dillow - have been saying this for a decade or so. Which is why it's so annoying to see a general acceptance of the claim that the profit share has been rising. People like Howard Reed (and someone you now are on a committee with, Richard Murphy) take it to be a proven fact.

    As you say, it ain't. The doubling of the VAT rate might have something to do with the fall in the labour share (which is I think what you're showing? Not the wage share? Or at least compensation of employees, not wages?) instead of profits.

    ReplyDelete

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.