7 Responses to “How to promote employment while protecting the low-paid”

Comments

1. I am not sure that the decline in the labour share can be attributed to globalisation. The labour share has not or only mildly declined in the US and the UK. The World Economic Outlook also reports a positive net effect of changes in trade prices (relative to domestic prices, which should measure external competitive pressures) on labour shares.

   Labour market flexibility achieved via dual-track strategies may have contributed to declines in wage shares more than globalisation. This would explain developments in Japan and Continental Europe. There are two mechanisms at work: i. new flexible contracts reduce the bargaining power of workers (fixed term contracts are indeed offered at a discount with respect to open- end contracts), and ii. there is a stock adjustment in the transition from a rigid to a dual labour market and new entrants in the labour market are less productive than incumbents.

   Finally, before providing policy recommendations, one should take into account that the labour share measures the functional distribution of income. Workers’ involvement in capital markets has been steadily increasing over time, with the development of pension funds even at relatively low incomes. This partly mitigates the distributional impact of declining wage shares.

   Yet, I agree that we need floors both in terms of minimum guaranteed income schemes for those losing jobs and wage insurance for the unskilled workers in the North suffering large wage cuts. Training helps too, but unfortunately dual labour market reforms are providing the right incentives. Training incidence among holders of fixed-term contracts is much lower than among permanent contracts.

   Posted by: Tito Boeri | April 12th, 2007 at 5:53 pm | Report this comment
   

2. Olivier Blanchard: A quick comment, based on a long worry.

   We (the economists) all agree that the trends in relative demand for skilled and unskilled workers probably require two types of measures from governments: a decrease in social contributions for low-skill jobs and the use of a negative income tax, to maintain both a low cost of low-skill labor for firms and the incentives for workers to take these jobs.

   What worries me is, if the trends continue (and I think they will), these measures will have a very large budgetary cost. This is already the case in France. The overall budget cost, the implied tax distortions, and political economy constraints, may well then make it unfeasible to extend or even continue these policies for very long (my guess is less than a decade). What shall we, the economists, propose then?

   Posted by: blanchard | April 12th, 2007 at 6:43 pm | Report this comment
   

3. Guillermo de la Dehesa: I think that Tito Boeri has raised a serious question about how the IMF has computed the labour share within total income using the functional distribution of income. The labour share can fall either because the total amount of labour income has fallen while income from capital has kept steady or because the latter has grown faster than the former, which is what it seems to have happened.
   To think that, in advanced economies, workers only leave out of wages and capitalists ony out of capital income is too simplistic . They both tend to live of both labour and capital or real assets incomes, although some more than others. So the fact that labour income has declined does not mean that workers incomes have diminished because their other financial and real incomes may have increased.
   There is no doubt that, in the las twenty years of accelerated globalization, capital income has been growing faster than labour income in most advanced countries. Companies have increased their profits and dividends, many have paid their executives with stocks and stock options, pension and mutual funds have increased their volume is an stagering way, there have been also several real estate and rents bubbles etc.
   It is also true that in most European economies total hours worked have kept falling and that in some of them unemployment has increased, so labour income may have decreased.
   The IMF states that labour policies in large European countries have contributed to a decline of the labour share. What seems to me a little surprising is to see that the IMF report shows that unemployment benefit replacement rates increases and not reducing tax wedges have been its main culprits, while empoyment protection legislation does not appeared to have had any effect. I had the idea that, as a result of globalization and technological change, in Anglo Saxon countries with more flexible labour markets the labour share could decline out of wage reductions and in countries with more rigid labour markets, high minimum wages, high firing costs and dual labour markets, the labour share would decline out of lower employment rates and increasing unemployment.
   Finally, the lower rate of growth of labour productivity in the large European countries over the last twenty years may have also affected to the labour income decline.

   Posted by: guillermo de la dehesa | April 13th, 2007 at 11:32 pm | Report this comment
   

4. Adrian Wood: There is a different (and potentially more illuminating) way of thinking about the share of labour in national income, namely as one minus the share of profits. The share of profits is by definition equal to the average rate of profit on capital multiplied by the aggregate capital-output ratio, which in turn is an average of the capital-output ratios in all sectors, weighted by sectoral shares of national income.

   Whatever has happened to the share of labour over the past couple of decades must thus reflect some combination of (a) changes in average profit rates, (b) changes in capital-output ratios within sectors, and (c) changes in the relative sizes of sectors with differing capital-output ratios. It would be interesting, as a follow up to the IMF report, to ask (i) what statistically was the accounting decomposition of the actual change in the share of labour into these three components, and (ii) what do we know or believe about the causes of change in each of these components.

   Without establishing the facts of (i), it is rash to speculate about (ii), but let me venture a few thoughts on each of the three components.

   (a) It is widely believed that profit rates have risen over the past couple of decades, reversing a downward movement between 1965 and 1985. But if so, then it is a bit of a puzzle, since real interest rates have declined during this period. Has the risk premium on investment risen? Or is the rate of profit determined in some entirely different way (Martin will remember a book I published on this in 1975).

   (b) Changes in capital-output ratios are likely to be partly a response to changes in profit rates (but if the elasticity of substitution were close to unity, then shares would not change), and partly a result of changes in technology.

   (c) If increases in trade have had an effect, it must have been mainly through changes in the relative sizes of different sectors. We know what trade has done to the composition of traded sectors classified by skill intensity: in developed countries, the more skill-intensive ones have become relatively larger. But for the purpose of thinking about the labour or profit share, one must also take account of what has happened in the non-traded part of the economy.

   Moreover, it is much less clear what one would expect trade to have done to the composition of sectors classified by capital intensity, which is what matters for the share of labour. Expansion of North-South trade should not have had much effect (as I argued in a 1994 book), but perhaps expansion of North-North trade contributed in some way.

   Posted by: Adrian Wood | April 14th, 2007 at 12:16 pm | Report this comment
   

5. Martin Wolf: I am grateful to Tito Boeri for bringing up the extraordinary and, in my view, noxious development of the dual labour market in much of continental Europe, with a sharp divergence between (usually older) permanent workers and (usually younger) temporary ones. I don’t know how much of the excess decline in the labour share in continental Europe this development explains.

   I see the point about the impact on training by firms. But temporary workers would seem to have a large incentive to invest in broad skills that give them relatively good alternatives in the job market. I agree with Tito on the distinction between functional and individual distribution of incomes. But I would guess that the main victims of the decline in labour shares – those with low skills – are also precisely those least likely to own significant financial or real assets. Indeed, I would expect the results for overall income distribution to be worse still, for this reason.

   Yet surely the big point is that the creation of the dual labour market, while increasing efficiency, is unjust. I cannot see how all those people who prate about the European social model can justify it. Surely, the terms need to converge, by making permanent contracts somewhat less so.

   I think Olivier has raised a huge issue. I have not run the numbers on this. But I imagine that if we try to support the incomes of the low skilled, while also moderating the marginal tax rates they face (or rather the rates of withdrawal of subsidies), then the total amount of redistribution would have to be very large and the marginal rates towards the top consequently high. I have been wondering whether possible ways out here are: first, tax bads more heavily; second, tax land more heavily; and, third, cut the supply of free public services and require co-payments by users. I agree that the combination of very large-scale redistribution with free government supply of health and education may be unsustainable in the long run.

   Guillermo’s comment fits in well with Tito’s. I have already discussed the distinction between the functional and individual distribution of income. I agree with him that some of the IMF results were pretty surprising. Such is the inscrutability of econometrics! Maybe the IMF authors will offer a comment on his queries about the impact of tax wedges and employment protection. But it is not clear to me that the lower rate of growth of labour productivity will affect the functional distribution. It depends both on what is determining that decline in labour productivity and on the characteristics of relevant production functions.

   Finally, Adrian raises some very interesting questions. It would certainly be useful to see the decompositions he mentions. My impression from the IMF analysis is that the big effect was from declining labour shares in unskilled-intensive sectors and declining shares of these sectors in the economy.

   Further, on his list of three points:

   (a) The divergence between rates of return and costs of funds is indeed astonishing. My view has been that it is best explained by the savings surplus condition of so much of the world, along with the low response of investment to high returns (which may indeed be a sort of risk premium).

   (b) I agree on this.

   (c) I have no idea what has happened to capital intensity (and profit shares) in non-tradeables. It is a good question. But trade might also have changed the labour intensity of all sectors. This might also have changed labour shares, depending on elasticities of substitution between labour and capital.

   I am surprised by the statement that it is unclear what effect one would expect trade to have on composition of sectors classified by capital intensity. I would have thought trade with developing countries has clearly reduced the relative size of the most labour-intensive sectors producing tradeable goods. Am I missing a point here?

   Posted by: Martin Wolf | April 17th, 2007 at 4:04 pm | Report this comment
   

6. Adrian Wood: To clarify my last point, which surprised Martin, trade with developing countries has indeed reduced the relative size of labour-intensive traded sectors in developed countries. The issue is what one means by ‘labour intensive’ - that is, intensive in labour relative to what other input? A central argument of my 1994 book, backed by evidence, was that what matters in the context of North-South trade is the input of (unskilled) labour relative to skill (or human capital), not relative to physical capital.

   In other words, and contrary to popular belief, developing countries are not at a comparative disadvantage with respect to physical-capital-intensive production (the main reason being that capital is so internationally mobile), but with respect to skill-intensive production (because most sorts of skilled labour are much less internationally mobile). For the full story, and the evidence, please see the book!

   Posted by: Adrian Wood | April 22nd, 2007 at 10:25 pm | Report this comment
   

7. Martin Wolf: I understand Adrian’s last point and agree.

   Posted by: Martin Wolf | April 23rd, 2007 at 12:54 pm | Report this comment
   

The FT Economists' Forum is a discussion among some of the world's top economists. As a general rule we accept comments from invited members only, but submissions from others will also be considered.

If you are a non-member submitting a comment, please include your relevant academic or financial background.

Post a comment

Name

Mail (will not be published)

As a final step before posting the comment, please type the two words you see in the image beloweight numbers in the audio clip; this test is to prevent automated robots from posting comments.

Type the two words: Type the eight numbers: Unfortunately you provided the wrong reCaptcha answer. Please try again: