Sunday, 6 September 2009

Past lessons, future policy

Michael Taft: With the news that the increase in unemployment is slowing down (though the range of missing numbers suggest that emigration may be rising at a considerable rate), let’s take a historical look at the last time Ireland emerged out of a recession with a high rate of unemployment. If unemployment tops out at between 14 percent and 16 percent over the next 18 months, how long will it take for unemployment to start falling?

In 1988 (the first year the internationally accepted ILO measurement was used) unemployment stood at 16.3 percent. By 1993 – with the Celtic Tiger growth ready to appear – unemployment remained stubbornly high at 15.7 percent, with the actual number of unemployed marginally higher than in 1988.

During this same period, annual GDP grew in volume terms by an average 4 percent. Employment, however, only grew by an annual average of 1.2 percent. Employment growth lagged considerably behind GDP growth.

The situation could have been much worse if we hadn’t benefitted from that ol’ standby – emigration. Between 1988 and 1993, over 100,000 had emigrated. Given that unemployment rose marginally in nominal terms during that period – from 217,000 to 220,000 – we can see what the effect would have been if people actually stayed in the land of their birth.

So, while GDP growth increased substantially, employment creation lagged behind and the only reason that the unemployment didn’t climb every higher was due to the economic safety valve of emigration. We should expect – and the IMF has warned everyone of this – that when GDP returns to growth sometime mid-to-late next year, unemployment may not start to deline for some time.

But there is one crucial factor we should be aware of during the late 1980s/early 1990s – something that is a bit of an embarrassment to the deflationists calling for massive public expenditure cuts; namely the role of the considerable stimulus expenditure engaged in by the government. During that period:

• Current expenditure increased by an average of 6.9 percent annually
• Capital expenditure increased by an average of 11.4 percent annually

In addition, during that period Ireland received another big stimulus in the form of European social and regional development funds. During the five-year period, this amounted €3.6 billion. This boosted public investment by 30 percent (in addition to the Government’s own public investment), and amounted to nearly 10 percent of our GNP in 1993.

Now compare that situation to what we are looking into over the next five years – severe cutbacks in current public expenditure coupled with a decimation of the capital budget. And all this without the benefit of EU investment funds.

Of course, we have to be careful in making comparisons between then and now. For instance, in the 1980s, the recession was relatively mild (GDP volume growth only contracted in one year). Agriculture played a more important role back then, while our export platform and infrastructural quality was relatively weak.

Still, a key issue which requires much more discussion is the role that increased public expenditure and investment played in eventually lowering unemployment, and its interaction with IDA policy, the devaluation and the emerging European single market. For while unemployment remained sluggishly high during the five year period we examined – starting in 1993, it fell quickly, from 15.7 percent to less than 7 percent in the following five year period.

Would this have happened without the massive stimulus the economy experienced? I would argue that it is doubtful. But what cannot be argued is the fact of that stimulus – something which the Dublin Consensus is in denial about.


Proposition Joe said...

You may want to also consider to rate of FDI over the periods 1987-1993 (while unemployment remained stubbornly high) and 1993-1998 (when unemployment fell dramatically).

Contrast with the recent trend giving us a much reduced share of global FDI, and the likelihood that competitiveness issues will limit our ability to reverse that trend.

Anonymous said...

The Governor-Designate of the CB posted (on an interesting mapping of unemployment(%) v. Current Balance as % of GNP since the 1960s on 17 July. The deterioration on both counts under the '82-'87 Govt. who left borrowing rip could be due for a reprise since we may be governed by a similar combo in the near future. All one is left with is fury at the unnecessary hardship and waste.

Sean said...

It might be worth considering that the economy of twenty-odd years ago was a very different creature than the one we have now - not only structurally in terms of worker-distribution over various sectors (you mention agriculture, but there's also manufacturing), but also comparitively in the European context, where we were much less advanced than other European countries then compared to now.

Bluntly speaking, it is much easier to stimulate an underdeveloped economy with simple increases in capital expenditure than it is with a more advanced economy.

Also, the state had a much more active role in the economy back then - we were still a "mixed" private/public economy. Now we are much more open. This to would drain a good deal of effectiveness from any stimulus.

Anonymous said...

Don't worry about all that. Sure there's massive scope for huge increases in borrowing and taxation of the well-off.

Mike Allen said...

The starting point for Michael's comment is the observed start in the turn in the rise of unemployment. A couple of thing are worth mentioning on that point.

Firstly the fact that there were three distinct peaks in live register unemployment in the late 1980s- mid 1990s. These reflected, as Michael notes, emigration patterns but also intermittent patterns of new jobs and complex patterns in who got the jobs. They also reflected changes in benefit rules. In January 1993 the CSO announced that the Live Register was over 300,000. Later they changed the way it was counted and the archive says it never reached that figure.

Which leads to the second point. To what extent is the current slowing down in increase the result in changes to Jobseekers Benefit rules? (higher eligiblity threshold and a reduction to 12 months). You might expect people who exhausted their JSB claim and then were ineligible for JSA payment because, say, of their partners income to keep signing for credits but this was by no means the historic pattern. I am not sure if the ramping up of training courses has been adequately taken into account.

It would be good news if it really is starting to turn (though limited as noted above) but I would not be too ready to assume the trend behind the dole figures just yet.

Michael Taft said...

Sean, there's much substance in what you say. However, I would point out that a considerable proportion of outlay in the stimulus programmes of other countries (e.g. US, Germany) has been on capital investment - either through the public programme or through the private capital base, as Germany is doing through renewables. Also, if we go beyond the 'bricks and mortar' image of investment we can see that in a complex economy other forms of investment are necessary. Notably, education but also other key aspects of our social infrastructure. For instance, though not formally part of the 'stimulus' package, President Obama's initiative on health care can be seen as an attempt to divert wasted spending in US healthcare into more prouctive areas.

Mike - your point regarding the rules change is absolutely correct. Indeed, we may be seeing the change to a two-year contribution record for benefit taking effect. There are a number of ways to manipulate Live Register figures (one thinks of Thatcher's diversion of unemployed on to disability). Therefore, it is important to read such numbers in conjunction with employment levels and the participation rate - which are in the quarterly national survey but which come out three months to six months late. We could see the perverse situation where real unemployment rises but the Government takes credit for a stailisation of the Live Register. It certainly wouldn't be the first time.