Few issues strike at the heart of the Conservative’s new electoral coalition like old-age benefits. They, too, have led to a reported split at the top of government with prime minister Boris Johnson on one side urging greater largesse and the chancellor Rishi Sunak on the other, keen to rein in further spending after the record borrowing through the pandemic.

In this case, neither is wholly wrong. Sunak is right to point out that there must be an end to giveaways or, at least, that they must be funded. But Johnson is also right to insist on pursuing his reform agenda, especially of Britain’s creaking social care system.

There are three issues facing the government. The first is whether or not to suspend the “triple lock”, which guarantees the state pension will increase by whichever is higher out of wage growth, inflation or 2.5 per cent. Due to pandemic-related distortions, average earnings are predicted to rise by around 8 per cent this year, meaning the commitment — included in the 2019 Conservative manifesto — would cost around £4bn more per year.

Second is to make good on Johnson’s promise to solve the crisis in social care. He is reported to be considering some of the proposals in the 2011 Dilnot report which called for a cap on individuals’ lifetime contribution to care costs. As only an unlucky few have to pay the highest amount, the state would share this risk and foot the bill for any costs over the cap. This would allow them to keep assets — including their home — to pass on to heirs.

The final puzzle is how to fund the other two. Options include reducing the tax-free “lifetime allowance” on total private pension contributions from over £1m to, say, £900,000, lowering the rate of relief enjoyed by higher earners, or taxing employer contributions more. All would hit traditional Tory voters, spooking the Conservatives after their loss of the Chesham and Amersham by-election.

Low government borrowing costs do not provide a route out of the dilemma. Increasing the state pension and implementing the Dilnot report would be permanent spending commitments, unlike the temporary pandemic support or the one-off spending on catch-up for schools. Sunak is therefore right that they ought to be paid for through permanent revenue-raising measures. Recent better than expected borrowing figures may persist, but it is unwise to count on it.

Funding spending on the retired by taxing those still saving is an unsatisfactory solution. The young are already likely to have far less generous pensions, especially compared with those on defined benefit schemes. Relative to a defined contribution scheme, a DB pension is less likely to fall foul of the “lifetime allowance” even if it guarantees a higher income. In the long term, reform is needed to ensure the young save enough.

Some restraint is needed too. While the triple lock was a manifesto commitment, it would be fair to suspend it this year. There is no reason why a temporary statistical distortion caused by the drop in wages during the pandemic should permanently raise old-age benefits. There is a better case for making the additional £20 a week uplift to the universal credit benefit permanent — in Britain, poverty is now more common among those of working age than the retired — which Sunak is keen to resist.

Implementing the Dilnot report, however, would be worthwhile. Britain’s social care system was badly exposed by the coronavirus and urgently needs a new funding settlement. The 2011 proposals are a decade old, but they remain a reasonable and fair formula for doing reform.