As a speaker at this summer’s Risk Management Conference in Muskoka (August 13 to 15) David Blake, Director of the Pensions Institute at Cass Business School in London and chairman of Square Mile Consultants, will talk about longevity risk. In advance of the conference, we asked David a few questions about the potential impact of longevity risk and ways to mitigate it. To find out more about David’s presentation and the Risk Management Conference, click here.
Who is affected by longevity risk?
It affects principally defined benefit plans; it affects defined contribution plans as well, but in a different way. But it’s a particularly important problem for mature defined benefit plans. By mature, we need to get the definition right here: we would call mature a plan in which the pension payments were larger than inflows from contributions and investment returns, but it might not necessarily be closed.
If you’re talking about a closed plan, in which there are no more accruals and no more people coming in to make contributions, then longevity risk has an even more severe impact.
How severe is the problem?
Well, one extra year of life expectancy at the age of 65 adds 3% to pension liabilities. So if your 65-year-old plan members were predicted to live so many years and they actually lived three years longer, that would actually add nearly 10% to your pension liability. It’s important and globally we have $25 trillion dollars of corporate pension plan liabilities facing these risks.
When do plans discover their longevity risk; plans usually have a three-year valuation cycle?
It would come up if the plan actuary, at the time of the valuation, decided to take a look at improvements in longevity. But often plan actuaries are not really skilled at that and they will simply use standard mortality tables and if those mortality tables are out of date – they are often updated only every 10 years – then you could easily get the liabilities being 7% to 9% bigger than the plan sponsor realizes. You have got to look at the actual mortality experience of the plan itself and see if this is out of line with that implied by the standard table. But as I say your average plan actuary is not very skilled at doing that.
What’s the solution?
Pension plans need to put on a longevity risk hedge. About 10 years ago, I recommended the government introduce longevity bonds. These would pay coupons linked to increases in the life expectancy of the pensioner population in the same way that the coupons on inflation-linked bonds increase if inflation increases. If pension plans could buy longevity bonds, this would provide a good hedge for longevity risk. However, no government has yet introduced longevity bonds.
Who’s going to bring them into existence?
Me, with a lot of help from other people! This is why I’m coming to Canada. I want to talk to people at this meeting to see the sense of doing this. Canada has not made the mistakes other countries have in terms of its plans yet. You like DB plans. You think they provide social cohesion, which they do indeed, and you’re prepared to stick with them. But if you’re not prepared to face the consequence of increasing longevity and do something about it, I’m afraid there will be big pressure on you to get rid of them, as has happened in the U.S., as has happened in the U.K.
Who would issue longevity bonds?
It has to be the government because only the government can engage in intergenerational risk risk sharing. Longevity risk is really an intergenerational risk, since the next generation pays for the pensions of the retired generation. The only agent in society that can initiate and enforce intergenerational contracts is government. No two private sector organizations could enforce a contract in which generations not yet born will be party to that contract. So it can’t be done in the private sector. It would have to be done in the public sector. The government is the insurer of last resort anyway. So if things go wrong, the government is going to have to bail out this generation’s pension plans. And that means the next generation of taxpayers will have to pay higher taxes to do this. Yet they are not receiving any insurance premium for offering this insurance. They would though if the government issued longevity bonds since there would be a longevity risk premium built into the price of the bonds.
How much would it cost the government to issue longevity bonds?
If retirement ages were increased in line with increases in life expectancy, there would be no increase in cost to the government. The government will actually receive a risk premium for issuing the bonds and for providing insurance: they would get an insurance premium. They’re actually providing this insurance without a premium right now. I think one day young people are going to get angry about this.
The premium is actually relatively small. We’ve done some simulations and show it’s between 20 basis points and 70 basis points per annum depending on how the bond is structured.
To learn more about the Risk Management Conference, please visit the conferences section of the CIR website. If you are interested in attending this event, please email Alison Webb to be considered, as limited space available.