Budget Responsibility and National Audit Bill [HL]

My Lords, my noble friend Lord Peston suggested to me that I should follow the introduction of this group of amendments by the noble Lord, Lord Higgins, by speaking now to Amendment 39. As noble Lords will be aware, this is simply an alternative means of achieving the objective that the noble Lord, Lord Higgins, seeks. One of the most important aspects of any piece of serious economic analysis is that it should be capable of being replicated. If the OBR’s forecasts are to achieve the status that we on this side and, I presume, the Government hope for them, they must be capable of being replicated. This can be done only if full information is available at the time of publication. The issue of replication is typically associated with the natural sciences, where replication of experiments is a fundamental requirement of any empirical scientific statement. However, the Minister may be unaware that it is now standard practice for any article published in a leading applied economics journal to provide the electronic address at which the data and other relevant information required to replicate the results in the article are available. In these days of large datasets and complex econometric models, data accessibility is critical to effective peer review—even effective assessment of whether any analysis or forecast should be taken seriously. Amendment 39, in my name and the names of my noble friends Lord Davies and Lord Myners, will ensure that effective appraisal of OBR forecasts and other economic analyses are possible. As is made clear in the preface to the OBR report that we discussed last week, compiling the fiscal forecast requires detailed information from many government departments. That is why our amendment refers not only to data and methods but to costings, which the OBR are required by the charter to confirm. In other words, all the raw materials on the basis of which judgments have been made and forecasts have been constructed should be available for objective assessments of those forecasts to be made. This will not involve any significant extra burden on the staff of the OBR, since the data and costings must already have been assembled in electronic form for the OBR to do its work. The noble Lord, Lord Higgins, raised an interesting point about the model that might be used by the OBR. We have been told that the Treasury will retain its own forecasting unit. We would like to know whether the forecasting model to be used by the this unit is to be the same as the model used by the OBR, in which case any differences in forecasts would simply be matters of judgment. That would surely be a ridiculous duplication. It would be much better to develop alternative perspectives, since they can often throw fresh light on difficult problems. In supporting the general line that the noble Lord, Lord Higgins, has taken, I simply add that we want to be in a position where serious researchers can replicate the approach and findings of the OBR in order to be able to evaluate them effectively.