Mr. Oisin O'Shaughnessy:

The first thing to say about the insights from the international comparisons is that it is really encouraging in that auto-enrolment works. That is what we see at an international level. It dramatically increases pension coverage and goes a long way towards pension adequacy. This links to Mr. Bolger's earlier comments that this is fundamentally the right direction and that has been proven internationally.

New Zealand was probably the first mover in terms of KiwiSaver and putting an auto-enrolment model in place. We sent a delegation from Irish Life to New Zealand to meet the different providers. To give a sense of the key insights and learnings, the proposed Irish model in this draft of the Bill is unusual by international standards. In New Zealand, inland revenue collects the premiums but that is all it does.

The money is then moved straight away to six providers that have been licensed to be AE providers in New Zealand. Because the New Zealand authorities have largely relied on the private sector, one key learning we can observe is that New Zealand was up and running in about one and a half years from the point at which it was decided to go for an AE model until that model went live. The UK model also leveraged a private market but it also incorporated a national employment savings trust, NEST, as a provider of AE, and stipulated that this would be a provider of last resort. This meant that if a worker did not want to use any of the existing providers, NEST would be there. NEST took four years to build because they built the full infrastructure. The lesson to be learned is that leveraging the existing providers led to it taking one and a half years to build in New Zealand and four in the UK. That is one of our key concerns. We all want this to be up and running, but it takes time when the existing is used. In New Zealand, the authorities supervise the six providers. Providers can fall off the list there and get onto it. We feel that there is quite strong supervision of the providers, and there are requirements. That can all be handled in terms of State concerns about providers. The key lesson around speed was one issue we observed as we looked over it.

There are also lessons in respect of complexity. People would need to be auto enrolled and inertia is a problem with pensions because the benefit is so deferred and feels so far away. One does not want 200 fund choices. The Swedish model took a long time to get to market. That model had a fund supermarket of 200 funds, which just confused people. For its pension holders, Irish Life would tend to want no more than about eight fund choices. We also have a default recommended one. A large percentage of our customer base uses the recommended one. In that way, we guide them through the right investment journey to retirement. Having a more limited fund shelf is a good idea, and this has been proven internationally. Our key point is that the market has this. The market innovates and improves on this over time. In this way, we feel that relying on the existing market, with the appropriate supervision, is the key learning as we look at it. Other jurisdictions such as the US and Australia rely on private providers as well. The Australian model has been up and running since the 1980s and has large assets under investment at this point. They are all largely relying on the private sector. The model that is probably the most different is that which obtains in Sweden. The Swedes had plenty of problems going live, and then probably took the wrong route on the fund offering.