David Willetts

In a speech to a Centre for Social Justice seminar at the LSE’s Centre for Social Exclusion, David Willetts explained how a neglected aspect of social justice is fairness between successive generations. This article argues that the large generation born immediately after World War 2 (the ‘baby boomers’) have benefitted from a favourable macro-economic environment throughout their lives, while the relatively small generation following them will bear the brunt of paying for the pensions and healthcare of their predecessors. Such extreme differences in the benefits and burdens of different generations over their lifecycles may need to be ameliorated in order to avoid a breakdown in the informal intergenerational social contract, which has sustained support for the welfare state over several decades. (This is a cleaned up transcript of a talk given to the Centre for Social Justice, and reprinted in the June 2007 issue of Benefits: The Journal of Poverty and Social Justice).

“One of the main functions of the welfare state is to redistribute income across the life cycle (Falkingham and Hills, 1995). You receive from the welfare state in your youth, contribute in middle age, and receive at the end of your life. Education is what we owe our children, and pensions and even health are what we owe our parents; “we” being the people in the middle who are by and large the working-age population paying the taxes. For me as a Conservative, I have always found it one of the most persuasive arguments for a welfare state. Capital markets are imperfect. A toddler cannot wander in to the bank and borrow the £50,000 that he or she needs for their education promising to repay it out of future earnings.

One way we could reform social security would be to do this redistribution across the life-cycle more simply and effectively than at the moment. We could apply Beveridge’s great insight that targeting and means-testing are not synonymous. Beveridge understood that if you identified the categories of recipients of benefits carefully enough, you could target without means-testing. Low incomes are concentrated by and large in particular groups, though it is not a perfect correlation. The poorest families are families with young children, just as the poorest pensioners tend to be the older pensioners. Simplifying the system with a higher rate of Child Benefit for young children and a higher pension for older pensioners would be quite well-targeted and could help to reduce the amount of means-testing.

This is a static model. The question we next need to face is whether a particular cohort is doing well out of a society’s economic and social arrangements. There is some literature on this, for example, David Thomson (1996) has written about ‘selfish generations’. Others have done elaborate calculations about what National Insurance contributions you have paid in during your lifetime, what you are getting back and whether there are particular generations that do well out of the introduction of the system.

A deeper question behind this is whether economic and social events over the past 10 or 20 years have advantaged a particular cohort. There are two examples of this phenomenon. The pensions crisis, certainly in the corporate sector, takes the form of companies closing their pension schemes to new entrants. At the same time they extract revenues from the company generated by the current workforce, including younger workers, in order to plug deficits in pensions schemes which are likely to be of benefit only for pensions of older workers. The younger generation are most unlikely to enjoy the kind of funded pension savings that older workers and current pensioners have.

The ownership of pension assets is going to be in the hands of specific cohorts and will not be transmitted to the younger generation. There is a very similar story in housing. People like myself who bought our houses in the 1980s, are sitting on a substantial capital gain. We are enjoying a significant increase in our wealth. But what about our children? Is the younger generation going to be able to get started on the housing ladder? They are going to have around their necks mortgages that are going to be long-lasting and expensive.

So when we look at the two most important forms of ownership of wealth in this country, pensions and housing, it is getting much harder for these forms of wealth to be passed on from generation to generation. I do not know if they are selfish or not, but there are certainly some lucky generations. This is not an accident; it is a result of some quite significant changes in the economic environment around us. I want to consider two in particular.

First, there is the massive distributional impact across the generations of moving to a world of low inflation. High inflation is extremely convenient if you are a borrower and low inflation is extremely convenient if you are a saver. In general, younger people are borrowers whilst older people are savers. The shift from high to low inflation has worked particularly to the advantage of my generation. We took out our mortgages when interest rates and inflation were high, the borrowing has been eroded by inflation and now we are protecting our savings by shifting to a world of low inflation. For the younger generation, their borrowings are not going to be eroded by inflation in the same way. The shift to low inflation is clearly having a generational impact on the distribution of wealth.

The second driver is the improvement in life expectancy. Those of us just above the cut-off age as the pension age starts rising are going to enjoy entitlements to state benefits starting at 60 or 65, or somewhere in between, which are going to be paid out to us for a very long time indeed. We are significant net beneficiaries. This applies to funded pensions as well. For the younger generation, by contrast, pension contracts are going to start to be adjusted. Twenty to thirty years ago when the monetarist controversy was at its height, monetarists used to try to get us to think about the impact on the economy of a helicopter dropping £10 notes and how this fed through both prices and the real economy. The new model that we have to think about is what if that helicopter was dropping pills that increased everyone’s life expectancy by 10 years. It is quite a close analogy with inflation because some of our contracts are denominated in terms of our age. The challenge for monetarists was to work out the distributional impacts of inflation when contracts are denominated in nominal terms. There are things you get an entitlement to when you reach a certain age and a society that has entitlements fixed by age and then has a significant shift in life expectancy is engaging in a big redistributive effort without realising or planning it.

So there is something quite significant going on here as a result of the combination of low inflation and greater life expectancy. The public choice theorists would tell us of course it is not an accident: the generation that are gaining by this are a very large, powerful generation. We should not be surprised that our nation’s economic and social arrangements seem to be working out in a way that particularly favours the baby-boomers.

We have got what the demographers would call a pig in the python. For Richard Easterlin (1980), you were at a disadvantage if you were in a large cohort because the competition within your cohort was greater, and that is indeed observable. But I would argue that the political and economic power of being a large cohort means that that cohort as a whole will do particularly well.

Imagine a country that balances its budgets over the economic cycle and which has stable public expenditure commitments. I do not mean by this that the amount of money is fixed, I mean that if there are more children you spend more on education, if you have fewer children you spend less on education, and if there are more pensioners you spend more, and for fewer pensions you spend less. Then think through the consequences of cohorts of changing size.

When the baby boomers are all children there is an increase in the education budget and other services for families with young children. Because you have a balanced budget rule and a relatively small working-age population the taxes paid by the working age population rise to finance services for the large group of baby boomer children. The baby boomers then arrive in work and the economy has a classic demographic bonus of a surge in its growth rate and a decline in pressures on public expenditure because you have got lots of payers-in and a relatively small number of receivers. You can lower tax rates whilst maintaining your balanced budget rule because you have got more workers and more tax receipts and lower expenditure.

The baby boomers then retire and require extra levels of public spending on pensions and the health service. Behind them there is a smaller cohort of the working age population who then face a double whammy: a smaller number of taxpayers and at the same time increases in public expenditure commitments, so taxes rise. Even with a balanced budget rule, the baby boomers will experience low tax rates when they are of working age and benefit from other generations paying higher taxes. That is why I think there is an issue of fairness across the generations, once you take into account the consequences of the different sizes of different generations.

One of the big issues in social justice, therefore, is social justice across the generations. How would we think rigorously about this? The issue at the moment is largely being developed by the environmentalists. The key arguments since Sir Nicholas Stern (2006) produced his excellent report, has been about this very issue; how should we price benefits and costs in the future? Stern’s model answers this question by breaking this question down into two smaller questions about future people.

The first question is to ask, given that they will be much richer than us, how much they will value changes in their income? When our descendants are all multimillionaires, how much will another 1% more income be worth to our richer descendants as against 1% of our income today? Nick Stern’s model suggests that the value of both should be the same. By selecting a value for what economists call the elasticity of substitution for consumption (or the ‘eta’) of 1, Stern has assumed that an extra 1% to spend now is equal in psychological value to an extra 1% to spend in the future (when we will be richer). For Stern, an extra thousand pounds for someone with ten thousand pounds will generate the same rise in happiness as an extra million pounds for someone with ten million pounds to spend.

Second, Stern has also considered how much of the burden for protecting that future income we should be willing to bear. He has done this through the social discount rate. Stern sets a very low social discount rate of 0.1% (effectively zero) in his report. He is saying that £1,000 in 50 years’ time is worth about as much as £1,000 today. If he had used a much higher social discount rate of, say, 2 or 3%, then costs or benefits out in 50 years’ time would be discounted down to pretty much nothing. Nick Stern’s social discount rate would, in fact, be zero out of moral principle. It is only 0.1% because, being the rigorous economist that he is, he wants to allow for the modest risk of the extinction of the human race. The 0.1% is to allow for the slight chance of there being nobody around to enjoy it. (This is in itself a little alarming. The Stern Review suggests it is 27,000 times more likely that the human race will be extinct than that I will win the lottery if I buy a lottery ticket every week for the next year).

Each of these assumptions has its own logic, and they are not incompatible.

To take an example, suppose we were considering whether to take action on a given problem where the cost was equivalent to 5% of income per capita in one hundred years’ time. Because of our eta value (1), we would say that even though the world in 100 years will be much richer, the loss of happiness caused by this problem would be exactly the same as a 5% loss of GDP would be today. (By contrast, if we had set a high eta, we could be saying that a 5% loss to those richer generations would only cause the same amount of unhappiness as, say, a 1% of GDP loss today and would scale down in present-day terms the cost of damage we think which we think will be done in the future).

Then, because of the effectively-zero social discount rate, when we are working out how much we should spend to prevent this catastrophe, we would be saying that the amount we would be willing to pay for abatement would be the same as the total cost. (By contrast, if we had set a higher social discount rate (say, 2%), we would only be willing to pay £138 to mitigate every £1,000 of damage in 100 years).

We could assume that we will get the same amount of utility from an extra 1% of income today as we will from an extra 1% in the future. We could also have a social discount rate close to zero so that we will be willing to meet the full costs of maintaining our descendants’ way of life andmaking sure they get the benefit of that 1%. However, a society which lived by these two assumptions would have very high savings (Martin Weitzman suggests 31%), because it would be willing to sacrifice a lot of money today to provide for the future.

However, this is not how we behave. We save very little, which reveals that we are not valuing the future in the way which Nick Stern’s model suggests. This is the key critique from DeLong, Nordhaus, Weitzman and Dasgupta. Stern has not just offered a neutral economic analysis; he has in effect called for us to attach more value to the future than we do at the moment. That is why he has not just moved from science to economics, but also to political economy. Behind the technical analysis there is a powerful moral appeal and Stern is trying to get to grips with a problem which Burke did not face; in a world of assumed constant growth, what does it mean to discharge one’s obligations across generations. If one assumes succeeding generations will be much richer, what does it mean to be intergenerationally just?

There is not much about this in political thought. One of the few authors who does wrestle with these intergenerational effects - though not with complete success – is John Rawls (1971). Rawls introduces the idea of the ‘original position’. He states that if our society were designed by a meeting of people who met in a scenario where none of them knew the circumstances of their own lives they would build a society which ensures that the least well-off
group (of which they might turn out to be a member) is as well-off as
possible

It is one feature of most social contract models, such as Rawls’, that contracting parties are contracting simultaneously. They all exist in time concurrently, so it is a horizontal contract with other members of your generation; it is not a vertical contract to successive generations. However, Rawls does believe that this group of people – drawn from across society and across time – should agree on a principle of intergenerational fairness. Unfortunately, he soon runs into difficulties.

Rawls wants to extricate people forming the social contract from anything that ties them down including culture or belief, or religion, or values, or language. One of the most profound and telling criticisms of Rawls, is that he is essentially saying that we might as well, for his purposes, be sheep. We are just bearers of DNA and there is nothing about our identities that we are allowed to take with us.

However, in the original version of his theory of justice, he describes the contracting parties as ‘heads of families’ and he says specifically that they have to reach their decisions about the social arrangements considering the interests of their children and grandchildren. This is clearly necessary for his theory in order to generate some kind of multi-generational account of where these obligations are coming from. We have to think about future generations and this is how he gets us to do it. But given that he will not even allow us to have language, given that he will not allow us to have a history or a culture, or a set of beliefs, or any values, why is he so keen to insist that we have children? It is a significant gap in his model that the only way that he can try to capture these obligations is by requiring the contracting parties to have children and grandchildren. In subsequent works (1993) he came to recognise this as a significant flaw in his theory and attempted to modify it such that the principle of intergenerational justice was that you should act as you wish generations previous to your own had acted. Hence he subtly changes his theory such that our intergenerational obligation comes not from our being potential parents, but rather because we have all been children.

This idea leads us to an alternative way to think about the problem of intergenerational justice which is derived from one of the great books that has influenced me, Robert Axelrod’s (1984) book, The Evolution of Co-operation. It is essentially trying to solve the prisoner’s dilemma. In America there are entire conventions where people gather as game theorists to model the prisoner’s dilemma – perhaps a slightly sad way for people to spend the weekend! The crucial insight of Axelrod’s book is that, if instead of thinking of this as a one-off, you think of it as an iterative game where you keep on replaying the prisoner’s dilemma, the correct strategy is not to betray unless and until you are betrayed. That is an extremely elegant solution. One of the reasons why I am a Conservative is that I value institutions and institutions are environments where people experience cooperation in exactly the form that Axelrod described. They are places where you learn to co-operate. That is one of the reasons why institutions matter to Conservatives.

The issue of intergenerational justice that I have been considering can be seen as a kind of slow motion Axelrod game across generations. This is similar to the ideas which have recently been explored by the game theorist Professor Ken Binmore. Each generation’s obligation is to sustain the intergenerational contract; provided that we care for our parents, our children will care for us, and so it goes on from generation to generation. If that is the nature of the obligation, then the danger with the current situation is that we may find we are breaking that intergenerational contract and are severing understandings that have sustained strong and effective welfare mechanisms over several generations. I am going to end with a quote from the thinker who does seem, to me, intuitively to have captured this understanding better than anyone else, Edmund Burke in his Reflections on the Revolution in France published in 1790:

‘Society is indeed a contract. Subordinate contracts for objects of mere occasional interest may be dissolved at leisure. But the state ought not to be considered as nothing better than a partnership agreement in a trade of pepper and coffee, calico or tobacco, or some other such low concern, to be taken up for a little temporary interest and to be dissolved by the fancy of parties. It is to be looked on with other reverence because it is not a partnership …..subservient only to the gross animal existence of a temporary and perishable nature. It is a partnership in all science, a partnership in all art, a partnership in every virtue and in all perfection. As the ends of such a partnership cannot be obtained in many generations it becomes a partnership not only between those who are living but between those who are living, those who are dead, and those who are to be born. Each contract of each particular state is but a pause in the great primeval contract as eternal society linking the lower with the higher natures, which holds together all physical and all moral natures each in their appointed place.’

Wrestling with this dilemma, as wrestling with so many others in social thought, leads us back to a very important Tory insight.

References:

Axelrod, R. (1984) The Evolution of Co-operation. New York: Basic Books.
Beckerman, W., Pasek, J. (2001) Justice, Posterity and the Environment, Oxford, Oxford University Press,
Binmore, K. (2005), Natural Justice, Oxford: OUP
Burke, E. (1790) Reflections on the Revolution in France. Edited by J. Clark (2001). Stanford: Stanford University Press.
Dasgupta, P. (2007), Commentary: The Stern Review’s Economics of Climate Change, National Institute Economic Review, 199, , pp 4-7
Easterlin, R. (1980) Birth and fortune: the impact of numbers on personal welfare. London: Great McIntyre.
Falkingham, J. and Hills, J. (1995) The dynamic of welfare : the welfare state and the life cycle. Hemel Hempstead: Prentice Hall.
Weitzman, M. (2007) The Stern Review on the Economics of Climate Change, Working Paper
Nordhaus. W. (2006) The Stern Review on the Economics of Climate Change, Working Paper
Rawls, J. (1971) A Theory of Justice. London: Methuen.
Rawls, J. (1993). Political Liberalism. New York: Columbia University Press.
Stern, N. (2006) The Economics of Climate Change: the Stern Review. Cambridge: Cambridge University Press.
Thompson, D. (1996) Selfish generations? How welfare states grow old. Cambridge: White Horse Press.