Semi-Daily Journal
Brad DeLong's Thoughts of the Moment on Economics, and on Other Topics as Well
DeLong's Home Page
September 25, 2002
The IMF Gets Less Optimistic
The IMF's latest forecast, as reported by the Economist, sees world economic growth as markedly slower than it had thought last spring. (But it does not call for further interest rate reductions and more stimulative policy in the U.S.: this surprises me, for the IMF seems to forecast growth slower than the rate of growth in potential output.)
It also blames slow productivity growth in Europe on Eurosclerosis--an attractive hypothesis, but one that I am not yet ready to embrace.
Economist.com: ...The IMF also appeared to back the views of the Fed’s chairman, Alan Greenspan, on the productivity improvements seen in America in the late 1990s. “The Fund reads the evidence as supporting the productivity boom,” said Kenneth Rogoff, the IMF’s chief economist on September 25th. But these improvements have not been seen in Europe, according to the IMF’s analysis, largely because of the failure to push through labour-market reforms, which has prevented Europe from benefiting as much as it should have from the hi-tech boom. This failure is particularly striking, in the IMF’s view, because until 1995 Europe’s productivity growth was ahead of America’s.
It is clear that continuing failure to address structural problems in Europe and Japan worries the IMF. Besides inflexible labour markets, Europe needs to tackle the problems caused by a rapidly ageing population. The IMF clearly thinks that the European Central Bank now has scope to cut interest rates. But monetary policy can only do so much, in Mr Rogoff’s view—structural reforms are needed too. And Japan’s economic performance over the past ten years, with its experience of persistent deflation unparalleled since 1945, has driven its economic partners to despair. Without far-reaching change, Mr Rogoff warned, there can be no guarantee against a “similarly bad decade”.
A time for caution
Sep 25th 2002
From The Economist Global Agenda
The International Monetary Fund has revised its forecasts for the world economy downwards. It says that, despite plunging stockmarkets and war jitters, growth should continue. But there are now more reasons to worry
A CHANGE of heart? The International Monetary Fund (IMF) probably wouldn’t go that far. In its influential forecast, the World Economic Outlook, published on Wednesday September 25th, the IMF says a global recovery is under way, and has been since late last year. But the IMF has made significant downward revisions to many of its forecasts compared with those published in April. And it concedes that there are more “downside” risks—ie, things could turn out worse rather than better—than existed just a few months ago.
Among the big industrial countries, the largest revisions are for Canada and the United States. The IMF now thinks Canada will grow this year by nearly a full percentage point faster than it estimated in April. Unfortunately, it has made almost as large a downward adjustment in its forecast for American growth in 2003. Canada may be the big success story of the G7 economies, but it is what happens in America, the world’s biggest economy, that matters.
The IMF is still expecting America’s economic recovery to continue, but at a slower pace than it previously thought. And with Japan just emerging from its third recession in a decade, and European economic sluggishness continuing to cause concern, global growth depends disproportionately on healthy American growth. Since the spring, America’s performance has been disappointing, and what the IMF calls “forward-looking indicators”—surveys of business and consumer confidence, for example—have fallen back markedly, in America and elsewhere.
The IMF’s assessment coincides quite closely with that of the Federal Reserve—America’s central bank—which met to review American interest rates on September 24th. The Fed decided against any further cut for the time being (though two members of the main policymaking body which takes the decision voted for a reduction); but it warned that the balance of risk continued to be weighted towards further economic weakness. The IMF warned that the Fed should be ready to cut rates further if the outlook deteriorates.
The IMF also appeared to back the views of the Fed’s chairman, Alan Greenspan, on the productivity improvements seen in America in the late 1990s. “The Fund reads the evidence as supporting the productivity boom,” said Kenneth Rogoff, the IMF’s chief economist on September 25th. But these improvements have not been seen in Europe, according to the IMF’s analysis, largely because of the failure to push through labour-market reforms, which has prevented Europe from benefiting as much as it should have from the hi-tech boom. This failure is particularly striking, in the IMF’s view, because until 1995 Europe’s productivity growth was ahead of America’s.
It is clear that continuing failure to address structural problems in Europe and Japan worries the IMF. Besides inflexible labour markets, Europe needs to tackle the problems caused by a rapidly ageing population. The IMF clearly thinks that the European Central Bank now has scope to cut interest rates. But monetary policy can only do so much, in Mr Rogoff’s view—structural reforms are needed too. And Japan’s economic performance over the past ten years, with its experience of persistent deflation unparalleled since 1945, has driven its economic partners to despair. Without far-reaching change, Mr Rogoff warned, there can be no guarantee against a “similarly bad decade”.
The IMF’s central assessment is hardly cheerful, especially compared with the optimism it—and many others—displayed only a couple of years ago. That optimism turned out to be misplaced. Yet if the current, more cautious projections turn out to be broadly accurate, they imply, for most countries, a slow recovery from last year’s global downturn. The forecasts, though, are subject to an unusual degree of uncertainty—with most of the risks on the downside.
Some of these risks are economic. The sharp fall in world stockmarkets in recent months will, if sustained, have an impact on growth, says the IMF. Its figures take into account the falls seen up to the first week in September. Since then, of course, the markets have taken a further battering, especially this week, fuelled by investor nervousness about accounting problems, corporate profits, the impact of higher oil prices, and, following Germany’s close election result on September 22nd, by concern that the re-elected government of Gerhard Schröder will fail to tackle much needed reforms in Europe’s largest economy. The actual impact on world growth is difficult to measure, says the IMF, but it is inevitable that the market declines will have some impact.
Another risk is that of deflation. Inflation in the industrial countries has fallen to levels unprecedented in modern times—1.4% this year, and likely to be 1.7% next. The IMF does not consider widespread deflation to be a major risk—“rumours of the death of inflation have been greatly exaggerated,” said Mr Rogoff. But he acknowledged that central banks need to be vigilant and ready to act.
More difficult to measure is the impact on economic performance of the current international political situation. If America invades Iraq, with or without the backing of the United Nations, nobody can make much of a useful guess about what economic impact this will have. Will it be a long conflict? Will there be wider political repercussions in the Middle East and beyond? Will relations between America and its principal allies be strained? The answers to these questions could alter the course of the world economy. A prolonged conflict, for instance, could push up the oil price sharply. The IMF reckons that a sustained rise of $15 a barrel—quite a modest assumption given what happened during the 1991 Gulf War—would cut global growth by 1%.
|
 |
Rogoff: a believer in boom |
In a global context such apparently small-sounding reductions are significant. They also serve to underline the extent of the collapse in Argentina, a country whose problems, while not likely to have much of an impact on the world economy as a whole, are greatly exercising the IMF. By the end of this year, the IMF estimates that Argentina’s economy will have contracted by a cumulative 20% in four years—that is twice the contraction experienced during the Great Depression of the 1930s, and unprecedented for any economy in modern times except for those involved in war or in transition from Communist rule.
The size of Argentina’s problems (the remedies for which are the subject of an increasingly bitter dispute between Argentina and the IMF) and the financial pressures which neighbouring Brazil now faces in the run-up to its presidential election have prompted the IMF to question the reliance of many emerging-market economies on capital inflows which, if they suddenly become outflows, can wreak havoc with economic policy and performance.
For developing countries as a whole, the future looks a little brighter—they might reasonably hope for 5.2% growth next year. And yet what happens in the industrial countries will directly affect them. The complexity of the problems now confronting the world economy are a clear sign that there is, as Mr Rogoff says, no elixir that will easily bring a return to sustained and higher growth.
Posted by DeLong at September 25, 2002 02:41 PM
| Trackback
One problem with structural reforms is that from an economic point of view, it's better to undertake them while the economy is strong, but from a political point of view, structural reforms don't seem to get attention until the state of the economy seems to call for them...
Besides, regarding the EU, I had in mind that some progress had been accomplished during the 90s on labour market liberalization. Also, the unemployment rate had been trending down until the after-shock of the US recession started to hit Germany, the UK and then the rest of the EU.
If I'm wrong, what I am missing?
I don't think the US productivity boom in the 90s was a mirage, but I have a hard time thinking that the productivity record would have been unaffected if there hadn't been a dot-com craze. I mean that a lot of investment money was chanelled to the IT sector and that can only have helped innovation in that sector (although there are certainly diminishing returns with this and bubbles do obviously come with a host of other concerns...)
Now, Europe didn't witness so much of a dot-com driven bubble, and so cannot be expected to have witnessed in the late 90s as much productivity increase than the US. However, I think Professor DeLong has made elsewhere (in the FT) a pretty convincing argument that the pool of IT innovation will inevitably reach and be applied to the other shore of the Atlantic, over time...
I'd note that the "structural reform" case is based on, as far as I can tell, exactly zero data points.
You can't count the USA as a success of structural reform, because it's always had the kind of labour market it currently has. The UK is a data point of a country which has changed from a European-style labour market to a US-style one, but its productivity is significantly lower than that of Europe. And Ireland went through its experiment with structural reform at the same time as it started to engage in destructive tax competition, while taking the advantages of being a small open economy in a currency union. I can't think of any other "structural reform" stories.
Daniel,
I have a question with regards to the UK, is productivity lower than Europe (not surprising), or is productivity growth lower (which would tend to support the structural-reform-doesn't-help hypothesis)?
There are, however, other candidates for structural reform "stories": Canada, the Netherlands, Australia, New Zealand and Spain (among others). In Canada, though, it is easy to miss the plot of the story: labour policy is controlled by the provinces, not the central government. There has been a shift to liberal policies in the past decade, but the change is coming slowly: Ontario and Alberta were early adopters, and now other provinces are following suit trying to copy their success. (Unfortunaly, the changes tend come with the whole right-wing package...)
On UK; productivity is lower; the gap appeared to be closing during the 1990s recession in Europe, but is now expanding again. UK closes some of the gap by working more hours than the Europeans, but that's not productivity ...
Don't know all that much about Canada, but the Netherlands retains a very European set of labour laws, and I'd suggest that Australia and New Zealand are too bloody volatile to be definitive. but, fair enough, good point.
Daniel
A short survey article on making productivity comparisons among G5 economies, published December 2001, is available on the UK official website at: http://www.statistics.gov.uk/cci/article.asp?id=84
David -
I seem to recall that througout the EU calls by economists to shorten the lenghth of full unemployment benefit eligibility and cuts in overall benefit generosity have been heeded, albeit partially and slowly.
I think some clumsy efforts have also been made to facilitate the hiring and disposal of young unemployed workers for short-term labor. Arguably, these efforts often constitude a bad patch against a background of very high administrative and economic costs associated with (hiring and) firing workers.
Any reference and comment would be greatly appreciated.
Oops, sorry! It's Daniel that I am responding to... :-7
Jean-Philippe
If you are looking for explanations for the relatively high unemployment and low employment rates in Eurozone economies, it is not just the eligibility for, and duration of unemployment benefits we need to look to but a whole range of other factors applying in labour markets, such as:
- statutory minimum wage rates relative to median or mean earnings
- statutory rights to maternity and paternity leave
- statutory consultation periods for redundancies and plant closures
- whether part-time workers have the same employment rights as full-time workers and after what qualifying period?
I don't know whether it still applies in Germany after some reforms of the benefits system but in the early 1990s it was reported that in some circumstances it was possible to attract more benefits income while remaining off-work sick than by returning to work. This was because sickness benefits related to previous earnings, including over-time and bonuses, before sick leave. Hence, if over-time prospects subsequently diminished it was apparently possible for sickness benefits to exceed likely earnings after returning to work.
There is a regular exchange of salvoes in UK media whenever a multinational company closes a manufacturing plant in Britain while keeping open a plant in mainland Europe. Some representative of the relevant union can be relied on to pop up to say: This has only happened because plant closure costs are so much lower in Britain and because minimal consultation is required. For balance, the media will then produce some representative of the governing classes who will: Yes, but Britain attracts such a high level of inward investment - as it does - because plant closure costs are relatively low.
It seems the concept of backward induction hasn't yet trickled down.
On one of the OECD websites I came across a whole collection of detailed reports on country benefits systems. I only browsed a few - the detail was truly daunting. Try: http://www.oecd.org/EN/statistics/0,,EN-statistics-0-nodirectorate-no-1-no-0-no-no-8,00.html
>>I seem to recall that througout the EU calls by economists to shorten the lenghth of full unemployment benefit eligibility and cuts in overall benefit generosity have been heeded, albeit partially and slowly.<<
Yeh, but this has happened over the last ten years and has not exactly been accompanied by a productivity miracle. The timing is all wrong, if you're going to take a rigourous approach to making statements about causation.
>>I'd note that the "structural reform" case is based on, as far as I can tell, exactly zero data points<<
I think this is overdoing it a bit. Take the following for example:
"We also find evidence that the impact of innovation activity (proxied by R&D expenditure) on productivity depends on market structure and technological characteristics, with a stronger
impact for technological leaders in high-tech industries. In addition, anti-competitive product market regulations are negatively associated with productivity performance. The negative effect is larger the further a country is from the technological frontier, because such regulations hinder the process of technology adoption. Finally, there is also evidence in the data of a negative impact of tight employment protection legislation on productivity when wages or internal training do not offset the higher adjustment costs associated with high firing costs."
This comes from the abstract of a piece of research by two European economists, Stefano Scarpetta and Thierry Tressel, for a piece of OECD research: "PRODUCTIVITY AND CONVERGENCE IN A PANEL OF OECD INDUSTRIES: DO REGULATIONS
AND INSTITUTIONS MATTER?". (September 2002).
Obviously they conclude, like I would, that they do. A roughly similar conclusion - in a piece by David Card and Richard Freeman - was highlighted by Brad in this blog. They concluded that changes in employment structures had been extremely effective in creating jobs in the UK.
The problem with the structural reforms debate is that 'every cook feels they can govern' and the argument has become polarised and sterile. Which is a pity because the well-being of a lot of people is at stake. The point here is not to make a checklist of the items which seem to make continental European societies more caring and then take out an axe to chop them away. At a time of declining birth rates legislation to support parenthood and child education would seem to be more than justified. However, and this is the point of the Card, Freeman argument, regulations which aid and abet the scandalously high levels of unemployment, and thus human misery, which France and Germany currently carry have to be misguided.
However, another problem with this debate is that really more than structural reforms, the continental Europeans, especially the Mediterranean countries are being asked to make cultural reforms in terms of flexibility in attitudes and approaches to risk. You see far more important than the firm level impediments to change are the attitudes to lifelong changes in occupation, mobility between regions (remember the EU needs this across countries), attitudes to risk taking and economic initiatives, and more relaxed and open attitudes to national identities and diversity. On all these the UK does decidedly better than its European neighbours.
Come on Brad, get off the fence.
No, your OECD study is miles away from supporting the conclusion you need:
>>Finally, there is also evidence in the data of a negative impact of tight employment protection legislation on productivity when wages or internal training do not offset the higher adjustment costs associated with high firing costs<<
Note that the clause after "when" basically sells the entire pass. You cannot view the internal training which is one of the key strengths of the French and German economies as being a separable phenomenon from the social and industrial model. Look at what's happened to firm-based training in the UK since 1980.
>>However, and this is the point of the Card, Freeman argument, regulations which aid and abet the scandalously high levels of unemployment, and thus human misery, which France and Germany currently carry have to be misguided.<<
Edward, this is quite high-falutin' rhetoric which is not supported by the data. The definitive work on this subject was an OECD study a few years ago, which concluded that there was no significant difference in unemployment rates across OECD countries which could be attributed to structural factors -- the business cycle, correctly allowed for, explained more or less everything.
The French and German rigidities did have an effect; they lengthened the average period of unemployment, and lowered the average age of the unemployed. But it is far from obvious that this can be used to support arguments based on "human misery". France and Germany, under this stylised representation, have greater stability of employment, while the US and UK have labour laws which make it much more likely that average workers will be fired and find another job. It is not obvious which system is more productive of misery, though a comparison of suicide rates and deaths from heart disease is suggestive ...
>>You see far more important than the firm level impediments to change are the attitudes to lifelong changes in occupation, mobility between regions (remember the EU needs this across countries), attitudes to risk taking and economic initiatives, and more relaxed and open attitudes to national identities and diversity. On all these the UK does decidedly better than its European neighbours. <<
Then why is productivity lower?
>Then why is productivity lower?<
Relating recent OECD studies at: http://www.oecd.org/EN/document/0,,EN-document-492-nodirectorate-no-4-24222-33,00.html
In the case of the UK, a persisting historic low savings/GDP ratio is part of the explanation and also, as the UK minister for schools recently put it, "It must be one of the most stunning statistics that we are 20th out of 24 OECD countries for staying-on rates [in education] at 17"
Absolutely. All of which says to me that labour market structure actually has very little to do with productivity (a concept in very dubious economic standing anyway), or with output. Guess I'm just a Keynesian at heart ...
>All of which says to me that labour market structure actually has very little to do with productivity (a concept in very dubious economic standing anyway), or with output. Guess I'm just a Keynesian at heart ...<
I'll settle for a compromise here even though I regard the Third Way as vacuous - both the demand-side and the supply-side matter.
What I cannot believe is that labour and product market structures don't matter at all and don't impact on the effectivenesss of fiscal or monetary measures to boost or curb levels of economic activity. Even Abba Lerner, thoroughly committed as he was to "functional financing" of public spending in his book: "Economics of Employment", certainly believed factor market structures mattered. I distinctly recall a passage - and this in a book published in the mid 1950s - where he argues that a fiscal stance intended to boost activity could be largely absorbed in pay increases for those already employed with little effect on aggregate employment or output. The issue has arisen very recently in another context in the UK with the increased public spending on education and healthcare.
As for productivity, I agree there are different measures, each with its own measurement problems and the measurement problems are the more acute in economies where 70% or so of output comprises services. Productivity comparisons between countries poses further problems from the choice of exchange rates applied to make the comparisons.
As for being Keynesian, there are so many flavours that the appellation need not convey much. After all, Pres Nixon claimed, "that we are all Keynesians now."
Perhaps the conceptual watershed is best exemplified in the kinds of explanations ventured to account for the persisting stagnation of Japan's economy since 1992, notwithstanding interest rates there reduced to zero and repeated fiscal boosts. All that is especially challenging for neo-classical or pre-Keynesian economics but not in terms of Keynes' General Theory with its focus on: "..an outstanding characteristic of the economic system in which we live that, whilst it is subject to severe fluctuations in respect of output and employment, it is not violently unstable. Indeed it seems capable of remaining in a chronic condition of sub-normal activity for a considerable period without any marked tendency either towards recovery or towards complete collapse." (GT p.249)
<>
Well for one, as they say, for reasons too numerous to mention here, but clearly historic ones have something to do with it. As Card and Freeman hold in their abstract:
"During the 1980s and 1990s Britain halted the relative declines in GDP per capita and labour productivity that had characterized earlier decades, and partially closed the gap in income per capita with France and Germany. These gains were mainly attributable to relative rises in
employment and hours. Unlike its EU competitors, Britain was able to achieve high employment-population rates with rising real wages for workers. The case that the change in economic performance can be credited to market-oriented reforms is harder to prove. Nevertheless, based on our own macro-level analyses, and micro-level evidence from several companion studies, we conclude that economic reforms contributed to halting the nearly century-long trend in relative economic decline of the UK relative to its historic competitors, Germany and France."
One interesting additional point on the UK productivity story that emerges from Card and Freeman is that the UK exited from agriculture much earlier than either France or Germany, and that both of these economies enjoyed a productivity 'spurt' during the 70's and 80's due precisely to the transition out of agriculture by one part of the workforce, a 'spurt' which was obviously not available to the UK.
However the UK's more rapid adoption - after all the UK has from the time of Hargreaves, Arkwright and Crompton been a land of 'early adopters' - of the ICT 'restructuring' may well now begin to give the edge again. Why is it that in the UK there is Vodaphone, whilst in the other major UK economies we have the ex state monopolies, or worse Vivendi and Bertelsmann. As I said this is about a cultural, as much as an economic transformation.
So I'm afraid it's a case of seeing what you want to see. The issue is a complex one, confounded as you rightly note by difficulties in measurement, but the evidence I feel is there that Britain has made a sea change, and in particular this is true of the ICT and knowledge-based end of things, areas which given the 'volatility' we're going through right now are hard to see clearly. However, beauty, as always, is in the eyes of the beholder.
>>The definitive work on this subject was an OECD study a few years ago, which concluded that there was no significant difference in unemployment rates across OECD countries which could be attributed to structural factors -- the business cycle, correctly allowed for, explained more or less everything.<<
I'm sorry Daniel, but no. See, eg, Olivier Blanchard's lectures available elsewhere on this site. Blanchard states:
"In a series of contributions, Steve Nickell (Nickell [1997], Nickell and Layard [1998])
showed that one could relate the increase in unemployment across countries to di®erences in labor market institutions. Following his lead, Justin Wolfers and I (Blanchard and Wolfers [2000]) looked at the panel data evidence;
we constructed measures of shocks and institutions, and concludedthat a specification based on observable shocks, and interactions of these shocks with measures of labor market institutions, gave a good statistical account of unemployment across time and countries over the last 30 years.We found in particular that differences in the response of unemployment to
shocks across countries could be statistically related to differences in institutions."
>>Guess I'm just a Keynesian at heart ...<<
Then it might just be worth noting that in a post-war treasury memo Keynes apparently noted that if "due to accidental bombing by a friendly power (the enemy axis powers were long gone at this stage EH) the industrial infrastructure of the North and North East of England were to be obliterated, with its management but absent its workforce, then nothing of great and lasting value would be lost"
Perhaps here is one key to some of Britain's historic productivity failings.
>>It is not obvious which system is more productive of misery, though a comparison of suicide rates and deaths from heart disease is suggestive ...<<
Could I recommend a perusal of Matt Ridley's "Genome" here ( a welcome relief I think from the heavy travails of Post Keynesian monetary theory) where you will find lots of interesting material about why there are different rates of suicide and heart disease across countries, including the surprising fact that having your cholesterol level too LOW puts you at higher risk of suicide due to serotonin deficiency, whilst being more independent and having more control over your working environment makes you LESS liable to heart attack due to having enhanced serotonin levels.
Edward, quoting Card and Freeman:
>>During the 1980s and 1990s Britain halted the relative declines in GDP per capita and labour productivity that had characterized earlier decades, and partially closed the gap in income per capita with France and Germany. These gains were mainly attributable to relative rises in
employment and hours<<
It is clear that you've been misled here; either someone has misquoted Card and Freeman for their own ideological purposes, or Card and Freeman actually believe that an increase in "employment and hours" could possibly change productivity. The input unit of labour is an hour, not a person!
>>In a series of contributions, Steve Nickell (Nickell [1997], Nickell and Layard [1998])
showed that one could relate the increase in unemployment across countries to di®erences in labor market institutions. Following his lead, Justin Wolfers and I (Blanchard and Wolfers [2000]) looked at the panel data evidence;
we constructed measures of shocks and institutions, and concludedthat a specification based on observable shocks, and interactions of these shocks with measures of labor market institutions, gave a good statistical account of unemployment across time and countries over the last 30 years.We found in particular that differences in the response of unemployment to
shocks across countries could be statistically related to differences in institutions."<<
This does not contradict anything I've said. I explicitly pointed out above that the average period in unemployment appeared to be affected by labour market institutions. What I said was that the overall, average level of unemployment, correcting for the business cycle, was invariant. All the work on impulse response functions in the world aren't going to prove or disprove that point. Please don't just cut and paste the first three returns from a google because they look as if they're related to a subject.
>>Could I recommend [...] <<
It was you that started bollocking on about "human misery", not me, mate. Could I recommend a quick perusal of the statistics on suicide, superimposed on a chart of the business cycle?
