The Labour and the Conservative approaches to welfare reform increasingly can be summarised as: "What would Beveridge do?" Seventy years after the Liberal reformer published his report on welfare policy, both parties lay claim to his mantle. The Work and Pensions Secretary, Iain Duncan Smith, insists that his quest "to make work pay" marks him out as the true heir to Beveridge; for Labour, his shadow, Liam Byrne, argues that a revival of the contributory principle is what the great man would have wanted.

But they should be wary of invoking Beveridge's name so readily - not only because doing so often masks a dearth of original thinking, but also because many of his insights are no longer applicable. Beveridge's vision of welfare was predicated on full employment, rather than an economy in which 23 people are chasing every new job. When Mr Byrne reminds us that Beveridge believed that unemployment benefit, after a certain period, should be "conditional upon attendance at a work or training centre", he should remember that he is not comparing like with like.

As Mr Duncan Smith has conceded, this is a "dreadful" time to be attempting welfare reform. George Osborne promised that private-sector job creation would "far outweigh" job losses in the public sector, but the number of public-sector jobs lost in the past year exceeds the number of private-sector jobs created. The coalition's vow to make work pay will ring hollow to those for whom there is no work.

Nor should the challenge of creating work for the unemployed obscure the related task of improving outcomes for the working poor. The welfare system was conceived by Beveridge as a form of social insurance, but it is being used more and more to compensate for falling real wages. A fact all too rarely reported in the media is that only one in eight housing benefit claimants is unemployed. The overwhelming majority are pensioners, or disabled people, or those working on low incomes, or those caring for a relative. The inflated housing benefit budget, which now totals £20bn, is the result of a conscious choice by successive governments to subsidise private landlords rather than invest in affordable social housing. Mr Byrne's simple assertion that the bill is "too high" is meaningless unless viewed in this context.

The coalition's pledge to raise the personal tax allowance to £10,000 is laudable, but this policy will do nothing for the three million households that earn too little to pay income tax, including many pensioners and parents who combine childcare with part-time work. A far bolder and redistributive measure would be to introduce a national living wage.

As the Labour peer Maurice Glasman argues on page 20, this would reduce families' dependence on tax credits and welfare and would represent "a change within the economy and not a transfer outside of it". As growth continues to falter, the question of how to distribute scarce resources will become even more important.

If Labour is to contribute meaningfully to this debate, it may need to rethink its stance on universal or "middle-class benefits". A commitment to universalism need not imply unconditional support for all universal benefits. A campaign urging the affluent to donate their winter fuel allowance to those in greatest need has raised £500,000, a reminder of how poorly targeted the payment is.

James Purnell, the former work and pensions secretary, has spoken of how resources could be diverted to fund a job guarantee scheme and a national salary insurance programme. If Labour is to win a hearing on this subject, it must do more than anxiously advance behind Beveridge's ghost.