Ireland has this pensions time bomb. It’s about demography: Ireland is a young country now, but in 2050 it will be old. The remaining young will have to look after the old. Looking after them will be costly so taxes will have to go up by quite a lot. Paying for care of the elderly will lower workers’ living standards a fair bit.

Part of it is literally about pensions, but more generally it’s about all the ways old people need support — in care homes, hospitals, and so on. 

Japan is where we’re headed. Japan is the oldest country in the world. In Japan, there are 2.4 children or elderly people for every worker. In Ireland, that number is currently just 1.1, though it’s rising every year. 

Made with Flourish

The government’s new auto-enrolment pension scheme is intended to help tackle the problem by nudging workers to put more of their own money aside for retirement. 

Stephen Kinsella wrote a detailed four-part series about the challenges of an ageing population. The summary is that a big group of middle-aged Irish people don’t own assets. They don’t own houses or pensions. When they age, taking care of them will be very costly. 

By way of solutions, Stephen suggested more housing, better technology, higher retirement age and more taxes. How much more taxes? Well, the median quintile of Nordic workers pays 27.7 per cent of their income in tax, compared to 19.2 per cent in Ireland. It seems to me taxes would have to go up by at least this much to care for the elderly, if not more.

It’s not that these ideas are bad – they are good. But the future sounds grim all the same. Working for longer with possibly stingier benefits and higher taxes on everyone else.

I think there might be a better way.

The scheme

The Irish state’s problem is similar to that of young Irish individuals. The state has big liabilities coming down the track in 20 to 30 years and it needs to put aside money now to cover them.

For individuals, the standard advice is to save for retirement. The way to save is to buy stocks, and over time, to gradually swap into something safer like bonds. This is because, in the long run, stocks return much more than other assets.

My crackpot scheme for solving the pensions time bomb is as follows: the state should borrow money from the capital markets, use it to buy stocks, and wait. That’s it. It should set up a leveraged sovereign wealth fund.

The basic idea is that, with great reliability, the stock market outperforms the bond market. And the Irish state can borrow money for the long term on unusually good terms. It can borrow for 30 years at 3.0 per cent.

Going back to 1928, the average return on the S&P500 is 9.5 per cent. The average return is higher if we choose any other starting year. In the ten years to 2022, the average return was 13.6 per cent, according to data from NYU’s Stern School of Business.

Say our new sovereign wealth fund borrowed €100 billion (€100 billion would increase our national debt by 52 per cent, but we’ll get to that). Like a worker, the goal is to save for retirement in 30 years’ time, so let’s say it allocated 100 per cent of the money, at first, to the S&P500 index of US stocks.

The annual coupon on the bonds would be €3 billion. Today the S&P500 yields 1.6 per cent in dividends, so there would be a shortfall. We’d have to sell down about €1.4 billion of the stocks to pay the coupon, initially.

But the beauty of the S&P 500 is that it has a strong tendency to go up over time. As we’ve seen, the long-term trend including dividends is 9.5 per cent per year. That would be our sovereign wealth fund’s tailwind.

The purpose of the sovereign wealth fund is to pay for the retirement of a generation of Irish people. So just like an individual saver, we wouldn’t want the fund to be held 100 per cent in stocks at the time we start to draw it down. We would want it to be held in a mixture of stocks and safe assets like bonds. Over time, we would gradually increase the proportion of bonds to, say, 60 per cent. Since 1928 the average return on US treasury bonds has been 4.9 per cent, but I’m willing to grant that might be more like 3 per cent over the next 30 years. Let’s assume 3.0 per cent returns on treasury bonds.

When we net out the coupon on the debt, the dividends and the capital growth on the stocks, and the return on the bonds, over 30 years, we end up with a pot of assets worth €744 billion. Repay the €100 billion bond and we have €644 billion in 2053.

What's €644 billion worth in today's money? If we assume inflation will average 2.0 per cent, €744 billion would be worth €358 billion in today's money. 

€358 billion is a lot of money. The entire Irish budget for 2023 will be €90 billion. 

Let's say we spent the money over 20 years, to help cover the costs an ageing society. That would give us €24.5 billion per year, again in today's money. Which is slightly more than the total health budget for 2023. 

Objection corner

What if the stock market went down? Over a 30-year time horizon, the risk of the stock market falling so much that we'd lose out is minuscule. Over the last 95 years, there's only one year in which the strategy would end up losing money over a 30-year horizon: 1928, the year before the Great Depression. Even then, it ends up losing just €10 billion in total. 

No country has ever done this. This is true. Though it's not a good reason not to do things in general. Plus, only so many countries can do this. Only small countries can buy lots of stocks without distorting the market. And only countries that can borrow cheaply could make the numbers work. Only a small minority of countries are both small and capable of borrowing at 3 per cent or less.

Wouldn't politicians spend the money before the 30 years were up? This is quite possible. Though it wouldn't be so different from borrowing money to spend it, which is what politicians are capable of doing now.

Wouldn't the fund get politicised and invested in politically popular but bad assets, which would screw up returns? This is the most salient objection. Would our sovereign wealth fund have the discipline to stay invested in a simple, proven asset like US stocks and bonds? To avoid losing money on the deal, we'd have to make about 4 per cent per year over 30 years.

Would the EU allow it? The stability and growth pact is a commitment by eurozone members not to run big deficits or borrow too much money. The borrowing limit is the one we might run up against — it's set at 60 per cent of GDP. But the Stability and Growth pact has been widely flaunted by EU members since the Global Financial Crisis and was suspended during the pandemic, and in any case, we'd be borrowing money to immediately buy financial assets, so maybe there'd be a way around this. 

Obviously borrowing €100 billion to play the stock market in contravention of EU rules is a bit of a crackpot idea. I'm willing to concede that.

But I can't think of very good reasons not to do it, either. The alternative does not look good.