For best experience please turn on javascript and use a modern browser!

Governments tend to postpone major interventions in pension systems. They are only prepared to adopt structural reforms when a major crisis occurs. UvA economists have come to this conclusion on the basis of a large-scale international study they have carried out.

‘Never waste a good crisis’, is a popular saying. And true enough, a crisis is the perfect time clean house. Fortunately, as it turns out, authorities take this message to heart. The downside is that they apply it to fix structural issues that have already been dragging on for a long time.

‘We always had the feeling that the pension debate underestimated the  importance of the business cycle’, says associate professor Ward Romp at the Amsterdam School of Economics (ASE) of the University of Amsterdam (UvA). ‘More detailed research has confirmed this. Despite lengthy negotiations, new permanent legislation is only introduced during the peaks or troughs of the business cycle.’

Romp is one of the four UvA economists who worked on a large-scale international study into pension reforms in the past few decades. Most of the work was done by Ron van Maurik who took stock of the pension reforms in all OESO countries and obtained his PhD on this subject last year. The other two economists involved in this project are professors Roel Beetsma and Franc Klaassen. The results will be published shortly.

State pension age

The reforms of the Dutch state pension law (AOW) illustrate very well how a government shapes its pension policy. ‘After the outbreak of the crisis in 2008, the Dutch government has addressed the AOW issue twice’, says Romp. ‘First, it agreed to raise the state pension age by 2020. Then, they accelerated the increase of the pension age in order to reach the age of 67 by 2021. At the same time, the future pension age was linked one-on-one to life expectancy.’ These were the years when the credit crisis was deepening, with property prices going down and unemployment rising fast.

In Romp’s view, the government actions were necessary to maintain an affordable state pension, but the measures were long overdue. ‘It proves that the government will only push through unpopular policies in truly difficult times. We had known since the 1980s that life expectancy would increase structurally. A link with the state pension age could have been made back then.’


The reasons that governments have their structural policies depend on the business cycle are mainly political. ‘The government sells its measures during a crisis by claiming that it is facing unpleasant surprises. The truth is that warnings about the consequences of certain demographic developments have been issued for years’, says Romp. ‘Moreover, solving the credit crisis and increasing the pension age are two completely separate things, as these measures will only lead to savings in the long term.’

The reverse is also true: at the peak of the business cycle governments become lax and tend to reverse earlier reforms. ‘This is what is happening now. The groundwork is being laid to reduce the speed of the future increase of the pension age.’ Incidentally, Romp believes that this could be a reasonable policy. ‘The current arrangement links the pension age directly to changes in life expectancy. The effect of this direct link is that the length of working life increases disproportionally to the length of the pension period, which is not needed to remain in control of the cost.’


The patterns found in the Netherlands turn out to be present in all OESO countries. ‘The results are robust as well as symmetrical. The effects emerge during economic downturns as well as economic upturns’, says Romp. ‘The data confirms our earlier suppositions on how reforms are implemented. Economic literature has come to refer to these reforms as “crisis induced reforms”’. Where pension reforms are concerned, we have now come up with solid evidence.

The dotcom crisis at the beginning of this century has, for a number of countries, marked a shift in their approach to the pension age. ‘A large number of countries have addressed the old-age pensions provided by the state. This was much-needed, because in many countries the state pension takes up a much larger part of the state budget than in the Netherlands, where private pension schemes play an important role.’ However, Romp notes that the measures taken by the Dutch government since 2012 have been more far-reaching than those in most other countries. ‘Most countries have set a path for future increases of the pension age that is static: the pension age has not been explicitly linked to life expectancy. As a consequence, unwelcome surprises remain a possibility, as life expectancy may increase faster than assumed earlier on.’

Learning curve

On the basis of the study, Romp concludes that governments pursue opportunistic long-term policies. One positive finding, however, is that governments adopt new measures more often than in the past. ‘Although most measures are implemented during economic peaks and troughs, we do observe that the number of structural restrictive measures has increased over the years. Possibly, the awareness of politicians has increased, as the views and the warnings of economists and institutions such as the IMF and the World Bank have been reaching them faster and because politicians have become more familiar with them.

Ideally, government policy is geared to respond to future shocks the moment the first rumbles are heard. ‘For instance, policy rules that include an automatic response to demographic changes would be perfect’, says Romp, ‘but unfortunately that is not how it works. At least, however, governments succeed in finding a time to take structural measures – that in itself is good news.’

More information? Email: