The Trussell Trust distributed its first food parcel in 2000. At the time, the network consisted of a single food bank in Salisbury, founded by a couple who had seen food banks operating in Canada and thought the model might be needed in Britain. By 2010, there were 35 food banks in the network. By 2015, there were over 400. Last year, the Trussell Trust alone distributed more than 3.1 million emergency food parcels. That figure does not include the hundreds of independent food banks operating outside the network.

The growth of food banks in Britain is one of the most significant social developments of the past quarter century. It is also one of the most contested. For some, food banks are evidence of a welfare system that has been deliberately and systematically dismantled — a consequence of austerity, of benefit cuts, of the introduction of Universal Credit and its notorious five-week wait. For others, they are evidence of the voluntary sector's capacity to respond to need, a demonstration of community solidarity rather than state failure.

Both interpretations contain truth, and the tension between them is itself revealing. The food bank has become a kind of Rorschach test for British political culture: what you see in it depends on what you already believe about the proper relationship between state, market and civil society.

What is harder to dispute is the profile of the people using them. Trussell Trust data consistently shows that the majority of food bank users are not homeless or unemployed in the traditional sense. They are people in work, or people receiving benefits, who have experienced a sudden financial shock — an unexpected bill, a benefit delay, a relationship breakdown — and have no buffer to absorb it. They are, in the language of social policy, the "just about managing": people who are, in normal times, getting by, and who are not getting by when normal times end.

The question of what to do about this is harder than either the left or the right tends to acknowledge. More generous benefits would help, but the evidence on benefit adequacy is complicated by questions of work incentives and fiscal sustainability. Better-designed Universal Credit would help, but the system's architecture reflects choices that are difficult to reverse without significant cost. What is clear is that a society in which millions of people require emergency food assistance in a non-emergency year is not a society that has solved the problem of poverty.