Wednesday, October 12, 2011

Wages, Labour and Output

I’ll be off travelling for the next couple of days, so don’t expect any posts until next week at the earliest. I’ll do my best to at least note any newsworthy economic news, but I might be too busy to keep up.

In the meantime, this post should serve to keep people occupied (and probably upset) until I get back.

In economic theory, the production of goods is considered to require a number of inputs including labour, capital and raw materials. Land could be considered as well, but I’m going to lump it under capital, because that’s how the data is categorised.

Under perfect competition and assuming a Cobb-Douglas production function with constant returns to scale, each input “earns” the value of the effort/cost that goes into making something. In other words, if labour contributes some portion of the value that goes into each product, its share of the revenue earned should be in proportion to that contribution. In technical terms, labour is supposed to earn its marginal product, and capital and the other inputs their marginal product.

In English: if output per worker increases, wages per worker should increase by the same ratio.

Mind you I’m going off memory here, so if my conception of this is completely off, feel free to let me know.

Anyway I tabulated the time series of output per worker (i.e. productivity) against wages per worker for the manufacturing sector (index numbers; 1959=100; sample 1959-2007; interpolated estimates for missing data years 1980 and 1998):


Conclusion: Since at least the Asian financial crisis of 1997-98, labour has not earned its marginal product. In English: wage increases have not kept pace with productivity for the last ten- fifteen years.

Here’s the log difference between the two:


I’m not going to flatly come out and say that manufacturing wages ought to be 40% higher, but…food for thought, yes?

Technical Notes:

Data on manufacturing sector output, wages and labour force from the Malaysia Economic Statistics – Time Series 2009 from the Department of Statistics


  1. bro selamat jalan...

    enjoy d non-smoking flight ya :P

    I suspect numbers are affected by the in-flows of foreign workers into the sector (can account up to 1/3 of total size)

    refer here (word file)

  2. Good one that - foreign worker influx was mainly in the agricultural sector before 1998, but increasingly in the manufacturing and other sectors after that.

  3. In 2008 -- latest year for which figures are easily available (EPU website) -- foreign workers accounted for about 10% of the manufacturing sector's total employment of 3.4 million. I don't think it's so much their presence in the manufacturing sector as their overall presence in the economy -- documented foreign workers, excluding maids, make up over 15% of the employed; including maids, it's close to 18% -- that's keeping wages down.

    That 30% figure in the attached word file is 30% of foreign workers, not 30% of manufacturing employment, unless we believe that another 20% of manufacturing employment is made up of undocumented foreign workers!

  4. the 1/3 is derived from 400+K foreign workers in d Sector Vs total employment numbers in specific sector around 1.2M

  5. This comment has been removed by the author.

  6. In the light of this revelation, please write on our competitiveness.

    (but have a safe trip back, kimqi nonetheless)

  7. @kj,

    Point taken. I used the manufacturing data because DOS issues monthly manufacturing statistics, which means the annual time series from where I sourced the data from could be updated. It might be worthwhile digging for service industry stats. Not too hopeful on that though.


    I was thinking about this on the plane ride, and I may have stumbled upon an alternative hypothesis. The discontinuity occurred in 1997-98, i.e. about the same time as the Asian Fianancial Crisis. Assume that this represented a psychological shock to both capital and labour.

    We know that one of the responses to the Crisis was a determined effort on the part of central banks to accumulate international reserves, as "insurance" against future sudden stop capital flight. The empirical evidence shows that they were right to do so - countries with higher reserves survived the great recession in much better shape than those that didn't.

    Extend that insurance motive at the micro level. Both companies and workers try to accumulate reserves (pay down debt, raise cash), but only companies managed to do so. The reason for this is that under a recession with higher unemployment, the relative bargaining power of capital is heightened.

    Hence profits are increasingly appropriated by owners of capital to serve as insurance against another crisis. Labour on the other hand trades off the ability to accumulate cash reserves against job security, at the price of being unable to demand appropriate wage increases. The ability of companies to secure low wage foreign workers reinforces the already poor bargaining position of labour.

    Does that make sense to any of you guys?

  8. From one angle, it does, hishamh - and thanks.

    But there's also the paucity of highly skilled workers; as one of probably many examples, local SMEs active as feeders to MNCs stationed here have griped that there are too few really skilled machinists who can take the imported parts to the next level in order to claim higher wages. So value-added crimped with the result wage growth slowed with no sign of changes in training programs, trainers and trainees. It then becomes a static MNC-dependent volume game and not higher local value from more precision and complexity. This is an overarching limiting factor.

    Similarly in services. Public sector wages go up in the hope that two years down the line private sector wages will follow. However this assumes the market will be buoyant but if it be otherwise, then decision-makers will look even more closely at automation which means if employment levels are to be sustained, services will have to also have more value-added quality which means better ideas or more measurable output. But ideas just can't be magic-wanded out of people and there's only so much output one can wring out of them.

    In all these discourses here and elsewhere, i am mindful we keep coming back to the human resource pool. Over the years i have tried to search for other development drivers. Maybe you have some ideas from what you've seen in other countries?


  9. Re relative power of capital and labour -- that's as much a political as an economic issue, and the power equation has likely just been further tilted in favour of capital with the recent Employment Act amendments. Labour is increasingly being pushed to give up not only on commensurate wage increments, but on security as well -- all under the label of "labour flexibility". Meanwhile, household debt increases.

    Re services, there are periodic censuses. Anyway, note that, outside the financial sector, per worker GDP in the services sector is considerably lower than in manufacturing, or even agricultural (put that down to commodity prices), and per worker GDP sets the ultimate limit to wages. As workers in services are even less empowered than the dismal state of workers in manufacturing, it's probable they get an even lower percentage of value-added. Question: will the planned shift to services as an engine of growth result in higher per capita GDP or GNI, simultaneously with a stagnation of even driving down of wages?

    Re human resources, granted that the quality of workers -- at all levels -- leaves something to be decided, but I'm continually puzzled by the incessant griping over unavailability of skilled workers and revealed demand in the ELX, with well over 2/3's of vacancies being for elementary workers (MASCO), when elementary workers apparently make up only some 12% of the employed.

  10. Walla, it always goes back to human beings and society. In the end, all policies should be geared towards that - what's the point otherwise? Yes, education and skill enhancement are the true long term factors. The problem is of course is that we have no way of really knowing progress in that area, expect through other means such as statistics, which capture reality only superficially. Moreover I think the pace of progress is so slow and diffuse, it will always be hard to discern - how to see the forest from the trees?

  11. @kj

    Great comments! Your question is particularly apt - here's what I think:

    I believe that we'll achieve high income status sooner rather than later, and the impetus will come from three avenues. First is actual real growth in output per worker but more important will be the increase in the ratio of the workforce to the population (note labour force growth is at least double population growth right now), and an appreciation in both real and nominal exchange rates. Most of the growth will occur from these last two factors - so you're right in that we could face a scenario with stagnant real wages, but a higher GNI per capita (though with a stronger exchange rate, imported consumption and durable goods should be cheaper, which may be construed as an increase in real welfare even if relative incomes and income inequality remain the same).

    The shift to services is important not because productivity or incomes are inherently higher, but because it creates a shift in the internal exchange rate between tradable and nontradable goods and services. A larger nontradable sector (higher internal exchange rate) supports a higher level of the external exchange rate, which raises the purchasing power of take-home pay, even if pay doesn't increase much in nominal local currency terms.

    That's why I've been saying the high income target is largely irrelevant. To me it's a done deal. I believe we'll become a high income economy sooner than we think. I also believe not many people will be very happy when we get there - it'll just be status quo ante, unless we pay more attention to long term structural factors and the distribution of income.

    Short to medium term, neither PR or BN are allocating enough attention to the current work force - the money set aside for retraining/reskilling is laughable. Education intervention costs (i.e. effectiveness of spending) increases with the age of the subject. Which means the best bang for the buck is actually at early levels of education. To be effective for the present workforce, you've got to throw serious money at upgrading skills - and we may not be able to afford that.

    Long term, the government is heading in the right direction - make pre-school compulsory. Second step - improve cognitive and social skill development at primary level.

    I don't know whether it may not already be too late for the current generation of workers, but we can still take steps for the generations ahead. But that's a hard slog with few "quick wins" to point to.

    No easy answers here I'm afraid.

  12. Hisham,

    Re tradeable and non-tradeable, point taken. However, with all this liberalisation of services what was non-tradeable becomes tradeable, and I fear a "race-to-the-bottom" for workers. Take health care services: what used to be a non-tradeable will become a tradeable, and costs to the user will rise faster than incomes; ditto education.

    Re education, true that returns to investment are much higher at early education than trying to compensate for deficits later in life. Indeed, it appears that the most important aspect of early education is not even so much cognitive, as non-cognitive social skills and attributes. While it is possible to compensate, somewhat, for cognitive deficits, it appears well-nigh impossible to compensate for non-cognitive deficits. That said, our problem is that the tertiary institutions turning out teachers for lower levels of education have become seriously deficient, and if we don't concurrently reform tertiary education, then investments in early education will be wasted. The same is true of many, if not most, of our so-called skills training institutions.

    As for whether we can afford it, perhaps the way to look at it is that we cannot afford not to afford it. Then perhaps we can have to will to see what can be cut and re-directed towards education and training, as well as how we can spend more effectively. Currently, the GTP on education appears to have adopted some of the worst features of so-called "reforms" in the UK and US, while the constant flip-flops means we never get around to proper evaluation, willy-nilly changing directions at the whims of the minister or whoever can command the ear of the minister.

    Re income distribution: given current policies and the socio-political climate locally and globally, there doesn't seem to be much prospect of serious reform. A "quick win" here would be raising the top marginal rates while simultaneously raising the level at which they kick in, a capital gains tax, and plugging some of the loopholes with regards to (sensitive issue!) zakat rebates. It is interesting to note that pre-tax, Norway's income gini is up in the low .4's, but post-tax, it drops to the high .2's.

  13. @kj

    It's a race to the top as much as to the bottom. Think of it this way - wage arbitrage in the nontradable sector is local, and wage arbitrage in the tradable sector is global. If global wages are higher relative to local wages for a given tradable industry, then equalising wages means local wages must go up. The converse is also true - lower global wages forces local wages down in the tradable sector.

    That's why there's been such an emphasis over the last decade on going up the value chain in exports. Higher value added means higher wages without having to muck about with the structure of the economy. Same principle with tradable services (Corollary: we're going to come up against higher inequality from all this.).

    Still, trade in services is still in its infancy globally. I think we're lucky in the sense that we're catching the tail end of restrictions on trade in services. There's still time to get wages adjusted before the world truly turns into a global village. 10-20 years from now we'd have missed our chance.

    With respect to education, I agree - it has to be total revamp. But the priority still has to be on the early years.

    As a matter of fact, what I had in mind in my comments before this was more about the present work force - those in their 20s and 30s with low skills and low ability to adapt to new skill requirements - rather than fresh grads. These are the people who will be the real losers over the next ten years as the economy transforms.

    On income distribution, again I agree - tax and transfer has been shown to work. But we need data and research on what the negative effects potentially are, particularly the impact of capital gains tax on investment. I don't know that I'd raise the top marginal income tax rate. I'd prefer setting both income and capital gains tax on equal footing - less loopholes, less gaming of the system, and potentially no loss in revenue (with the caveat that I haven't done any serious research on this subject yet).

    Another avenue is seriously promoting profit sharing schemes (no, not share options) as a way of equalising capital income across different income levels.

    And hey, seriously, thanks for your comments. You're making me think rather deeply on these issues, which is fun.