Facebook Blogging

Edward Hugh has a lively and enjoyable Facebook community where he publishes frequent breaking news economics links and short updates. If you would like to receive these updates on a regular basis and join the debate please invite Edward as a friend by clicking the Facebook link at the top of the right sidebar.

Wednesday, November 22, 2006

The Japanese Labour Market

In what follows I intend to step back from the precipice a bit, and take a rather longer term look at one feature of Japan's economic performance, its labour and employment market. In this sense this post is going to form part of a "tripple whammy" I am working on, where I will attempt to carry out an in depth examination of possible connections and interconnections which may exist between population ageing, rapid job creation and weak internal consumption in three key G7 economies: Italy, Germany (see here) and Japan.

Rising-Employment Falling-Consumption?

Well lets start by looking at the story so far, at least as far as Japan goes. Fortunately we do have a number of key stylised facts at our disposal. First off, unemployment has been steadily - I could say relentlessly - dropping in Japan since reaching a peak in 2002.

And since the start of 2006, we have seen this decrease continue almost month by month until the market seems to have turned, sometime around July 2007. Since that point it would seem that conditions in the Japanese labour market have been steadily worsening.

and new jobs have been created, lots of them:

Yet at the same time domestic demand has not been revived to anything like the extent that might have been expected. Retail sales have been in decline for most of the last year as you can see from this retail-sales purchasing-managers-index for Italy (remember that on the PMI reading, anything below 50 represents a contraction).

Meanwhile private household consumption, despite some early strengthening, could hardly be said to have been booming. We did see a couple of (in Italian terms) comparatively strong quarters in the first half of 2007, but then, surprisingly - as can be seen in the chart below - the rate of increase began to weaken in Q3, while, curiuosly, in the position in the Italian labour market remained, more or less, stable.

So how can we account for this apparent paradox of a steadily tightening labour market and deteriorating internal consumer demand? One explanation for this could, of course, be that labour market performance is a lagged indicator (that is that it only registers economic deterioration after other indicators have been pointing to red for some time, and this is surely true), but could there not be something more going on here? Especially since this kind of pattern is being repeated in Germany and Japan, and these two countries have, along with Italy, the highest global median age, and as a result the highest proportions of potential workers in the older age groups. There is something very different about the labour market tightening we are seeing in Italy, Japan and Germany, for example, and the kind of inflation generating labour market tightening which we are observing in Eastern Europe. So why this difference?

Italy's Age Structure

But first off let's take a look at one of the younger age groups for a minute, the 15 to 24 one. As is well known this group is now in historic decline as a proportion of the total Italian population. The decline has been very rapid, with a drop of around one third (from 15% to 10% of the population) since 1990.

What this means, logically enough, is that there are steadily less and less people in this age group to fill places in the labour market. To this numerical decline we need to add theongoing secular decline in economic activity rates among this group, as more and more of Italy's - now scarce resource - young people delay entry and seek to improve their education, their human capital rating and hence their future earning capacity.

Thus the proportion of this group which is economically active has been declining steadily, as have the absolute numbers of those who are active, and the numbers of those who are actually employed. It is perhaps worth noting that the absolute size of this age group has been virtually stationary over the last 3 or 4 years (a statistical effect), but it is now set to fall steadily.

So if new employees will be hard to come by in this age group as we move forward, where can employment growth come from? Well basically there are two evident potential sources of labour, immigration and older workers. It is hard to envisage any large increase in employment in the 25 to 34 or the 35 to 54 age groups since - as can be seen from the chart below - activity rates among these groups are already fairly high, and even the slight fall-off which can be seen to have taken place recently in the 25 to 34 age group seems to be the result of a decline in female activity rates, and this is almost to be hoped for if Italy is to do one thing which is very important for its long term future, and that is have more children. Squeezing this particular lemon too hard at this point in time is only likely to obtain short term benefit in return for substantial negative long term outcomes.

Turning now to the principle sources of potential long run labour supply, in the first place it is obvious enough that there has been a significant surge in the size of the immigrant workforce in recent years (see chart below, and my other post for more details).

The other main potential source of additional labour in an ageing economy is the over 55 age group (and as we move forward of course increasingly it will become the over 65 one). Now many international agencies (the World Bank, the OECD, the IMF, the EU Commission etc) are pinning their hopes on the idea that the effects of population ageing may be to some extent offset by increasing the participation rates of these older workers, presenting impressive looking projections all the way out to 2050 to back their view that this is a workable solution. Yet since we have, in the here and now, a number of examples of societies who are trying quite hard to follow recommendations here, I do think it is important to examine in detail what is actually happening in these societies and try to really start to estimate the longer term macroeconomic consequences of this shift.

Now the important thing to bear in mind when we speak of older workers is that the important decision they need to take is about whether or not to continue working, and what is very clear in the Italian context is that workers in the 55 to 64 age group are increasingly taking the decision to stay at work, in some form or another.

Not only is the activity rate - which has, it must be said, been ridiculously low for this group, especially given Italy's very high life expectancy level - on the way up, the unemployment rate is on its way down, and the number of those employed is steadily rising.

More Work, Less Pay?

What is also significant about this trend is that it is also associated with a significant growth in part time and temporary work. The question really is who is doing this part-time/temporary work? In Japan it has become clear that many people now leave their "lifelong" job at 55, only to continue working in some way shape or form for another 15 years or so (both the Economist and the Financial Times have recently run articles about this trend in Japan - see here - although they fail to explicitly lock-it-in to the ageing population issue, which I think is where it belongs) . The work ethic in Japan is probably quite different from the one in Italy, but the similarities in the way the labour markets are evolving are really quite striking . In the German context it is also clear that the growth in part time and in non-social-security covered employment has been significant. And of course, in Germany, as I explain at considerable length here, the big increase in employment is in comparatively low skill, low wage work, which very often draws the over 55s into employment in much the same way and for much the same reasons as it does in Italy and Japan.

So what IS observable in the case of all 3 of these economies is that they are experiencing at one and the same time skill-shortages in some key areas of economic activity (due to the growing population crunch among the young - Japan for example is notoriously short of nurses) and generating large volume employment in more tenuous and lower skilled categories of work, since such work meets the skill and performance profile of the workforce they really have available. Thus, even as these labour markets tighten we find NO real evidence of significant wage-squeeze push.

From this point in we are left guessing until someone does some really systematic research, but it isn't a bad guess to suggest that the pressure on wages in the more skilled areas is being offset by a downward movement in wages in those less skilled areas where a mixture of lower skilled migrants, retirees and older workers are offering themselves for work in increasing numbers.

One other conjecture about Italy is that migrants are doing jobs in Italy which retired or semi-retired workers are doing in Germany and Japan, and hence we find that the participation rates in the 55 to 64 age group are still pretty low. At the moment I'm not quite sure what the macro economic consequences of this are going to be.

My feeling is that since people reaching 60 can now expect to live quite a long time, and since nowadays there is no great certainty attached to current levels of retirement benefit as we move forward, then older people, who are normally more prudent, will be protecting their current savings - in whatever form they may hold them - as best they can, and supplementing pensions with bits and pieces of work to maintain living standards so as to not run down their capital.

Another point here is that many retirees in Italy have a lot less in the way of accumulated wealth (and imagine the situation in a country like Hungary, which is the next one coming in this group as far as I can see, even though the median age is somewhat younger, but the male life expectancy is also much lower, and the population is already falling) in comparison with Germany and Japan.

This is why the recent deal Prodi struck with the Italian unions about postponing raising the retirement age was such a negative when viewed from where I am sitting.

So what I am suggesting is that the very weak internal consumption we are seeing in these three countries (and I would drop-in that Hungary is "coupling" here perfectly with the others, in terms of the model I am working on) is not ONLY associated with a higher propensity to save associated with older people, but also to do with the earnings profile associated with the new kinds of low level work older people are doing. In other words you can't just take the large number of new jobs being created and translate this into more consumption (as I think most of the conventional analysts are doing) since more things are happening here.

North-South Regional Stresses and Imbalances

But in any event the data we have is fascinating. The regional disequilibrium in Italy seem to be once more becoming really important (just like East-West one in Germany, and Tokyo vs the rest in Japan). While the national participation rate for the 55 to 64 age group went up from 28.9% in Q1 2004 to 33% in Q3 2007, in the mezzogiorno it has gone up from 31.8 to 35.3 over the samer period, so the South is keeping pace here, but if we look at the 65 plus group, while participation has gone from 3.4% to 4% nationally over the same period, in the mezzogiorno it has gone DOWN from 2.4 to 2.1%. The 15 to 64 participation rate also dropped from 54.1 to 52.5 over the period in the mezzogiorno while in the North it went up from 67.8 to 69.2 %. And this situation is reflected in the relative job creation performance between the North and the South.

Basically, given the very strong fiscal pressure which is about to come in Italy, and the danger IMHO of a sovereign default at some point if nothing is done to correct this very weak growth trajectory, Italy can be almost literally torn apart by this disequilibrium, especially given that it is reinforced by the unequal distribution of migrants. We have an ongoing polarisation of wealth, employment and people, and we really aren't giving sufficient consideration to the longer term political implications of the underlying economo-demographic process.

New Forms of Employment: Temporary and Part-Time Work

I have also found a limited breakdown for part time work by age. The two categories which the Italian statistics office use are "15 to 34" and "35 and over". Now strange as it may seem the number of part-time jobs for the 15 to 34 age group actually went DOWN between Q1 2004 and Q3 2007 - from 1.107 millions to 1.102 million - while among the over 35s it went up from 1.74 to 2.121 million. So Italy's new part-time workers are by-and-large not young, and it is a good bet that the majority of these new workers come from the over 55 group, and that it this kind of work which is responsible for the increase in the participation rates at the higher ages.

Of course, when we come to look at TEMPORARY work the pattern is rather different, there are an increasing number of young people (and since the number of such people is steadily declining, a rising proportion) working on temporary contracts. The number has gone up from 1.035 million in Q1 2004 to 1.368 million in Q3 2007. Over 35s (which we can pretty much imagine as over 55s, since the 35 to 55 age group is normally pretty robust in employment participation terms) goes up from 679,000 to 993,000.

By Way of a Conclusion

Basically the macro economics of all this are hard to assess. Italy's working age population - ex migration - has touched the ceiling, and without immigration it will go down and down. So everything depends on raising the productivity of those employed. But raising productivity today is pretty much synonymous with raising the human capital component and if in volume terms the numbers of older but less qualified people working - and working in more and more fragile and less and less well-paid occupations - swamps the number of new highly educated workers in highly productive jobs (we are talking about aggregates here) then the new value created by the society in question won't compensate for the contraction in the workforce. This is particularly true when it comes to raising participation rates in that oft quoted potential labour supply, female workers over 55. Many of the women in question are excellent wives and mothers, but given their often very low level of formal education, and given their lack of real experience of work out of the home, the economic worth in value added terms of their formal labour market participation may be much lower than many expect, and certainly this is where the evidence to date is leading us.

I also feel that the Italian experience is very similar and comparable with what we have been seeing in Japan and Germany, so it seems to me that there is now strong prima facie evidence that we need a big and really systematic research programme into the details of all of this, and rather less of that "gung-ho", we haven't got a problem approach, which has prevailed up to now, and which - at the end of the day - is based on the idea that raising participation rates will do the trick. As we are seeing, and unfortunately, it may well not do. It will do something, but that something may well not be enough.

Saturday, November 18, 2006

Stephen Roach Interviews Franco Modigliani

A Conversation with Franco Modigliani

I am reproducing here an interview which Morgan Stanley economist Stephen Roach had with the late Italian Nobel economist Franco Modigliani. The interview was originally published on the MS Global Economic Forum in October 2002, but is unfortunately no longer available online on that site.

Stephen Roach (SR)

There are always great imponderables in the macro outlook. But today the questions seem to loom larger than ever. At the top of my list are two burning issues -- the prognosis for the American consumer and the risk of a US deflation. Nobel laureate Franco Modigliani has long been noted for his path-breaking work on these very issues. He was in town in couple of weeks ago, and we had the chance to sit down for a most engaging discussion. An edited version of our conversation follows.

Let's start out with the burning issue of the day -- the American consumer. Franco, as a pioneer in the macro of consumer behavior, what is your assessment of the outlook for this key sector of the US economy?

Franco Modigliani (FM)

The consumer is at risk. Having said that, I must confess to being surprised about the resilience of consumption after the stock market bubble popped. I would have expected some retrenchment. I would have also expected some related damage in the real estate market and the dollar. I still believe at some point there must be some repercussions.

What gives you such conviction?

The key for the consumer response is not the immediate situation. Too much is made of the day-to-day developments, such as mortgage refinancing. My work shows that consumers frame decisions in the context of their life-cycle considerations. It is an intertemporal theory, whereby decisions in any period have implications for adjustments in the future. That's why the wealth effect can be important. If the popping of the equity bubble results in a reduction of longer-term return expectations on an important asset, that has important implications for life-cycle saving and spending patterns.

But there is more to household wealth than stocks. Hasn't the property market saved the day?

While it may seem that way right now, I have my doubts. I am suspicious of those studies that find the wealth effect is larger from real estate than equities. Theory tells me it should actually be the opposite. That's because the house in part, produces a consumer good -- housing services, which we consume. When the value of the house I inhabit goes up, its implied rental value increases. But that does not significantly improve my spending power, because my imputed rent has gone up as much. Any wealth effect on individually-owned property must net out the consumption of the service we derive from living in our homes. Those adjustments need not be made for stock portfolios. It is possible that new refinancing instruments, such as home equity loans may have temporarily distorted this relationship. But I would view this as a one-time shift, not as a permanent realignment of the link between wealth and consumption.

So how does it all end for the American consumer?

A negative wealth effect tells me it can't go on forever. And that's when I revert to the life-cycle theory. Sadly, the large cohort of aging baby boomers is not adequately prepared for old age. The personal saving rate is too low. It has been depleted by individuals betting on asset markets. Life-cycle theory suggests that the saving rate should have gone up by now. Obviously, it hasn't -- at least, not yet. While that puzzles me, it doesn't dissuade me from the basic view that the balance between consumption and saving will have to adjust. It's just a matter of when. That leads me to conclude that that the American consumer is the most dangerous portion of the picture.

Let's turn to another key issue being debated in financial markets -- deflation. I find myself more worried about deflation than at any point in my professional career. Are you at all concerned about the possibility of deflation?

I was surprised when I first heard of the extent of your concerns. They come across as very strong, very scary. I think you are probably exaggerating.

I have been known to do that.

Does Byron agree with you?

No, nobody does. And that is one of the reasons why I am trying to argue the case -- to give some legitimacy to the deflation debate. These days, when you merely utter the word "deflation," you get strange looks like you're crazy.

Well, let me start by saying one thing at the outset: It really depends on how much deflation we are speaking about. I would say that deflation is something to worry about, only if is going to be large. If it's just a temporary blip down in the price level, it wouldn't worry me. It has to have both depth and duration to be a serious concern. But so far, we really haven't had any deflation at all in the United States. Isn't that right?

So far, you are absolutely right. But the GDP price index is now increasing by just 1.1%. That's the lowest rate of aggregate inflation in the US economy in 48 years. So we're not that far away from the hallowed ground of price stability, to say nothing of the deflation that lurks on the other side of this threshold. But it's not just the data. I worry about deflation for both cyclical and structural reasons.

In other words, you are not just worried about the existence of deflation. You are more worried about it having bad effects.


Well, that takes us to a consideration of debt deflation. Essentially you will agree that inflation or deflation has an effect both on creditors and debtors. In that sense it is a zero-sum game -- whatever is gained is lost and vice versa. But the private sector as a whole gains because it is the net creditor of the Great Debtor, namely the government. In practice, firms are also net debtors and creditors are mainly (older) individuals. So from the household's point of view, you might argue that deflation is good. However, I think it is a very important thing to remember is that deflation is bad for the government -- as a net debtor; in effect, it increase the real national debt. On balance, that is good for the current generation that owns the claim on government indebtedness but bad for future generations that will have to pay higher real taxes to finance that claim.

Can I just ask you to clear up one thing: Are you saying that creditors gain because deflation effectively expands their purchasing power?

Yes, it's both the purchasing power with respect to the flow -- debt service -- and the purchasing power of the stock of assets. While retired people lose out from the standpoint of interest earnings in a deflationary period, when they liquidate their capital, they can exchange the proceeds for more goods. In that respect, the stock effect works to their advantage in a deflationary period.

I'm very interested in this differential effect. It seems to unmask some of the tensions that we macro practitioners lose sight of. Can you elaborate on this point?

Yes, your question actually takes me back to a story about a good friend of mine, Harvard economics professor John Kenneth Galbraith. Well, one year, when I was a visiting professor at Harvard, Ken Galbraith invited me for a debate with the students at Lowell House. He decided that we should have a debate about inflation. He made the point that inflation was bad for the poor and good for the rich. And I said, "Ken, you're way off." [Laughter] The poor, I insisted, are more likely to be debtors. And debtors benefit from inflation because of the reduced real burden of the debt. So I concluded that inflation is good for the poor and not for the rich.

And what did he say?

[Laughter] I think it was at that point that Ken said to the students, "Don't believe what he says. He's one of these people who writes textbooks and has to defend what he said in his textbook." But there's a moral to this story: In assessing the macro impacts of inflation, deflation, or fluctuations in interest rates, the distribution of indebtedness matters. For every borrower who gets a boost in purchasing power, there is a lender who loses. They may be different people, but it is the net effect that matters for the macro economy.

Let me go down a slightly different road. Deflation has an impact, as you just described, on the ability to service debt. But deflation has also become a real wake-up call for companies that are coming to realize that they don't have any control over pricing leverage. And they're cutting costs in response. If there comes a point when the cost cutting is aimed at headcount and wage payments, how does an overly indebted society adjust to that?

The answer to that question depends on the causes of the supposed deflation, a subject on which you have not dwelt extensively. I can think of two major causes: The most obvious would be a deep cyclical decline, but this version is most unlikely because of the availability of well known stabilization policies, and because in this century, as a result of wage rigidity, no contraction has ever produced deflation in developed countries, since the Great Depression. The other possibility is dogged foreign competition from countries like China. This type of deflation, within limits, would be good for the consumers and would compensate for any resulting downward pressure on nominal wages.

How might such "good" deflation turn into bad deflation?

The only problem I could see would be the possibility of debtors getting worse off. And even though there may be offset by creditors being better off, that doesn't help if you have wholesale repossession of houses and other levered assets. You also suggest that the price pressure could reach a point where it could lead to a curtailment of domestic production. However, I doubt that even this kind of deflation could get very far because it probably would be offset by a devaluation of the dollar.

So if it's debt deflation that has you most concerned, don't you worry about record levels of household and corporate indebtedness currently existing in the United States?

Macro debt loads have certainly gone up a lot -- it's hard to deny that. But, here again, I want to look beyond the macro and study the composition of debt. And there are some very recent developments on the scene that give me pause -- namely the home equity loan. I think this new instrument has done some very important things. Suddenly, a house has become a liquid asset.

Do you think that's good?

Well, from the standpoint of progress, I think it's good. I have always been in favor of making assets more liquid. In that same vein, I am also in favor of allowing loans on 401(k)s.

But the flip side of that liquidity is debt -- the means by which incremental purchasing power is extracted from these assets. Does this create a new bias toward debt-financed consumption that can only end in tears?

That's obviously a worry, especially if you are right on deflation. What Americans do now is borrow against future asset values. Deflation would bring down the value of the properties themselves and therefore reduce the ability to borrow. This phenomenon can become very important if deflation deepens and lasts.

I have one last question on the deflation issue: Doesn't the endgame hinge on wage rigidities, or the lack thereof? If deflation pushes nominal wages down, won't all bets be off?

As a Keynesian, I believe that macro variables such as employment and prices hinge on wage rigidity. Once you accept that, all the rest follows. Historically the floor for increases in wages is defined by productivity. So as long as productivity growth is maintained, nominal wages are unlikely to fall. It's a very deep question. Employers simply don't like the idea of going to workers and telling them they're going to cut their wages. And because employers rely on the support and loyalty of workers, they can't afford to alienate them. By and large, prices are linked to such "sticky" wages. In short, Steve, your case for deflation is interesting. But I'm not persuaded that that the wage-setting mechanism is flexible enough to turn this into a serious problem here in America.

Thank you very much, Franco. I certainly share your concerns on the consumer. And I will continue to probe the case for deflation.