The Cruncher (February) is wrong to conclude that the [UK's] N[ational] H[ealth] S[service] comes out poorly just because British five-year cancer survival rates are lower than in other countries. An alternative explanation is that Britain does not waste money on extensive testing procedures for diseases it cannot cure. For such diseases, and many cancers are among them, earlier diagnosis merely serves to raise the number of years between the identification of the disease and death—it does not affect expected mortality. Peter SugarmanThis illustrates the danger of relying too heavily on seemingly appropriate but overly narrow indicators, whether under the current regime of micro-managed Mickey Mouse indicators devised by bureaucrats, or under a Social Policy Bond regime. We need, as a society, to think very clearly about the role of government and what we want our taxpayer contributions to achieve. These decisions today are largely made by lobby groups, coporate interests, politicians and bureaucrats whose over-riding goal is to retain power. A Social Policy Bond regime would express its objectives in terms of meaningful outcomes, which would be discussed and refined rather than, as now, falling out of long, complex, inaccessible, adminstrative procedures.