Baroness Altmann wants to abolish the triple lock on pensions. She's wrong.

First, let's dispose of a myth - that the triple lock is yet another of those policies that benefit the old at the expense of the young. It doesn't. If it remains in place, it will in fact be of greatest value to today's youngsters. This is because they will benefit from decades of growth, whereas today's oldsters will see only a few years of it. And remember - the power of compound growth is very great.

Instead, the issue is a different one: how should we provide for our pensions? Should we do so as individuals by saving, or through the state via higher future taxes? The triple lock is, in effect, a form of creeping nationalization which shifts the job of pension provision from the individual to the state because the higher is the state pension, the less we need to save for ourselves.

In this context, it is nonsense to say that the triple lock is expensive. Given low annuity rates, any form of pension income is expensive. Yes, it'll be expensive for the state to provide a high future pension. But it'll be expensive for young people today to get a future pension through their savings. The issue is not the expense, but how that expense should be incurred. Should it be by the state or by the individual?

Several things make me think it should be the state.

One is that private pensions incur big management charges - and these compound horribly over time. Cynics might think this is why the Baroness is opposed to the triple lock: she's thinking of protecting the interests of her prospective future employers.

Secondly, the job of saving for yourself requires you to solve problems which are almost insuperable. On the one hand there are irrationalities such as the present bias which cause us to save too little. But on the other hand, it's also possible to be irrationally prudent and to save too much. A state pension takes the impossible job of intertemporal choice out of the individual's hands.

Thirdly, individuals face huge long-term investment risk. Even if the economy grows steadily, the stock market might not: studies show very little correlation over the long-term between economic growth and equity returns. This could be because of distribution risk - the benefits of growth might go to workers rather than shareholders. Or it might be because economic growth accrues to firms which don't yet exist. Because the government can fund future pensions out of future taxes, it doesn't face these risks.

In fact, there's another way in which the government is better able to bear risk.

Real bond yields are negative now and the market expects them to remain so for years. This might well be because investors expect low economic growth. If this is the case, then anyone trying to save for a pension faces a massive problem - that returns on financial assets generally will be poor. There's a risk that this problem could get worse - that trend growth might fall even more, further depressing real returns for savers.

But the government can bear this risk better than savers. If real yields fall further, it will be rewarded even more handsomely for borrowing. What's a horrible danger for savers is thus no problem for the government. This too argues for the government to take on the job of providing future pensions.

You might object that lower trend growth will be a problem for the government because lower future GDP will make it harder to raise the taxes to pay higher pensions. But lower future GDP is also a problem for pension savers, as it means lower dividends on shares. The issue isn't how to avoid this problem, but rather who is better placed to cope with it.

These factors make me suspect that the government is better than the individual at providing future pensions. Insofar as the triple lock is a form of gradual nationalization of pensions savings, it is therefore a good thing. It is almost certainly George Osborne's greatest legacy as Chancellor.