To maximise wellbeing, important public policies and programmes must work. They should be rigorously evaluated to ensure that they have the intended effect, are efficient, and meet their objectives. If not, such programmes should be scaled back and certainly not extended.

In recent months there have been a few changes to KiwiSaver and more are under consideration. These relate to KiwiSaver default providers, default funds and contribution rates.

But those who are inclined to tinker with KiwiSaver really should be looking at the evidence on its efficacy and thinking about whether or not any changes could help KiwiSaver meet its objectives.

To help us do that now, here’s a quick refresher on the purpose of KiwiSaver. As outlined in the KiwiSaver Act (2006) it is to “encourage a long-term savings habit and asset accumulation by individuals who are not in a position to enjoy standards of living in retirement similar to those in pre-retirement” – the target population. Further, KiwiSaver aims to “increase individuals’ well-being and financial independence, particularly in retirement, and to provide retirement benefits.”

This brings me to the recent changes that were announced to KiwiSaver default providers and default funds, which are reviewed from time to time. The number of default providers has been reduced from nine to six, the risk profile of default funds has changed from conservative to balanced and the Government will now ensure that savings in default funds are invested ‘more responsibly’.

While the first two of those changes are not particularly concerning and may even be beneficial for some KiwiSaver members, the Government deciding what ‘responsible’ investment is, and is not, is a little more troubling.

In this particular case it means excluding any investments in fossil fuel production. This, we are told, reflects the Government's commitment to address the effects of climate change and transition to a low-emissions economy.

But addressing the impacts of climate change is not the objective of KiwiSaver and pursuing multiple objectives in this way makes it less likely that KiwiSaver can actually do what was intended. It also misses the fact that we have alternative, and much better, policies to address climate change. In particular, the Emissions Trading Scheme (ETS), which sets a limit on total emissions and puts a price on those emissions, has been in New Zealand for almost as long as KiwiSaver. The ETS has recently been made even more effective and is the most efficient way of reducing emissions, as Green Party Co-leader James Shaw pointed out last year.

Beyond this, where will the Government’s prescription for ‘responsible’ investment end and what will it cost us? Will we next be required to invest in local infrastructure, electric vehicle production, low flow shower heads and social housing? Will investing in companies that manufacture utes and barbecues or the wrong type of food be banned next?

We’ve also recently learned that the Government is investigating a proposal to increase KiwiSaver contributions. New members, and those who opt into it, would see their contributions increase from 3 per cent to 10 per cent in half a per cent a year increments.

This is surprising because unlike most public policies, KiwiSaver has actually been rigorously evaluated against its stated objectives. In fact, to the credit of the scheme’s architects, KiwiSaver’s evaluation was a requirement when it was introduced. KiwiSaver was found to be a poor performer, so calls to expand the scheme are rather puzzling.

I worked on that evaluation for several years, undertaking two major studies which were later published in peer-reviewed journals.

The first study by Law, Meehan and Scobie used a survey designed specifically to evaluate KiwiSaver. The second by Law and Scobie used administrative data from Inland Revenue matched to the Survey of Family, Income and Employment (SoFIE) which followed a large group of individuals over an eight-year period. The beauty of using SoFIE was that it measured changes in individuals’ assets and liabilities both before and after the introduction of KiwiSaver.

The first study found that only a third of the contributions that individuals made to their KiwiSaver accounts were new savings, while two thirds would have happened without the scheme. This was due to substitution between KiwiSaver and other forms of saving which would have occurred anyway (such as paying down a mortgage).

What matters even more are effects on wealth but using different data and techniques the second study showed KiwiSaver membership had no effect on net wealth accumulation on average. That’s even after controlling for other wealth accumulation factors such as income, age, gender, ethnicity, family circumstances, home ownership and previous levels of wealth.

The studies also looked at the efficiency with which KiwiSaver reached those it was intended to help. Again, KiwiSaver performed poorly. For every success, where KiwiSaver helped someone with an apparent savings problem to save more than they otherwise would have, fourteen other KiwiSaver members were merely along for the ride. This much larger group either did not change their saving behaviour or had no savings problem in the first place. KiwiSaver is so inefficient at reaching its target population that each success costs around $300,000 in subsidies.

At this point proponents of KiwiSaver usually complain that since these results are a few years old now they cannot be relied upon. Indeed, they said that from the day they were first published (analysis of data takes time). Yet, in the intervening years these KiwiSaver proponents have been unable to provide any rigorous evidence showing KiwiSaver has been successful at meeting its objectives, nor make even a credible argument as to why one would expect things to be different now.

In the meantime, the international evidence base on the efficacy of schemes like KiwiSaver keeps growing. It turns out our experience is not unique. There are many examples, but the one everyone now points to is the work by Harvard Professor Raj Chetty and his co-authors. They examined 41 million observations on savings for the population of Denmark and found that each additional dollar of government expenditure on subsidies increased total saving by only one cent.

The Government has made a lot of the importance of following science and evidence in their response to Covid-19 of late. It would be great to see more attention paid to evidence in other areas of public policy.