Many environmental choices have effects that last far longer than the people who make them. Carbon dioxide stays in the atmosphere for centuries, some nuclear waste remains dangerous for thousands of years, and an extinct species never returns. So ethics has to ask: what do we owe to people who do not yet exist?

Reasons to care

The Irish-born statesman Edmund Burke wrote in 1790 that society is a partnership “between those who are living, those who are dead, and those who are to be born.” Many traditions share this idea of a chain of generations, each receiving an inheritance and passing it on. Philosophers give several arguments for obligations to future people:

• They will be real. Future people will feel pain, need food and value clean air just as we do.
• Time is like distance. Many argue that when someone is born is as morally arbitrary as where they are born. We would not ignore harm to a stranger because she lives far away; why ignore it because she lives far in the future?
• We have the power. Our choices shape the world they inherit, and they have no voice in those choices.

The 1987 United Nations report Our Common Future, often called the Brundtland Report, gave the most widely cited definition of sustainable development: in paraphrase, development that meets the needs of the present without compromising the ability of future generations to meet their own needs.

Puzzle one: discounting

Economists comparing costs and benefits at different times usually apply a discount rate, counting future amounts as worth less today. Some reasons are widely accepted: future people will probably be richer, so an extra dollar matters less to them; the future is uncertain; and money invested today can grow. More controversial is pure time preference: counting future well-being as less important simply because it comes later. The British philosopher and economist Frank Ramsey called this, in 1928, a practice that is “ethically indefensible and arises merely from the weakness of the imagination.”

The rate chosen matters enormously over long periods. At a 5 percent annual rate, a harm costing $100 a century from now counts as about 76 cents today; at 1 percent, about $37. Climate debates have turned partly on this: the 2006 Stern Review used a very low rate and called for strong early action, while the economist William Nordhaus used higher, market-based rates and favored a more gradual path. Both supported action; they disagreed on its pace.

Puzzle two: the non-identity problem

The British philosopher Derek Parfit described a deeper puzzle in Reasons and Persons (1984). Which particular people are born depends on countless details, including when their parents met and conceived them. A large policy choice, such as whether a country uses up its resources quickly, changes those details for millions of families. So the people living a century later under one policy would be different people from those living under the other.

Now suppose we choose depletion, and the people born later have harder, but still worthwhile, lives. Were they harmed? They are not worse off than they would otherwise have been, because otherwise they would not have existed at all. Yet the choice seems clearly wrong. This is the non-identity problem. Common responses say that an outcome can be worse even if no particular person is worse off, or that whoever comes to exist has a claim to decent conditions.

Giving the future a voice

Some governments have created institutions to speak for future generations, such as Hungary's ombudsman for future generations (since 2008) and the Future Generations Commissioner for Wales, created under a 2015 Welsh law to check whether public bodies consider long-term effects.

Worked example: the cheap liner

Imagine a county choosing between two designs for a new landfill. Design A uses a standard liner that costs $2 million and is expected to last about 50 years. Design B uses a thicker, double liner that costs $5 million and should last 200 years. If the cheaper liner fails, leaking chemicals could contaminate groundwater; engineers estimate the cleanup would cost around $40 million, roughly 80 years from now.

Discounting. At a 5 percent discount rate, $40 million in 80 years counts as about $800,000 today. On that calculation Design A wins easily: why spend an extra $3 million now to avoid what counts as less than $1 million? At a 1 percent rate, the same future cost counts as about $18 million, and Design B looks like the better deal. The choice of rate decides the answer.

Ethical questions beneath the numbers. Is it fair that residents in 80 years, who get no benefit from today's cheap disposal, bear the cost? Does a dollar figure capture what it means for families to lose clean well water? And who should pay: the future county, or the people who produced the waste?

Non-identity. Someone might argue that future residents are not harmed, since different people will live there whatever we choose. Most people find this unconvincing as a reason to choose the cheap liner, which is exactly what makes Parfit's puzzle a puzzle.

A sustainability test. The Brundtland definition asks whether Design A meets present needs by compromising the ability of future residents to meet theirs. Many would say it does.