The Office of Budget responsibility is a recent invention. George Osborne wanted a body that was said to be independent that could assess government economic policy and set out in forecasts what the results were likely to be. To do this he transferred the Treasury model for forecasting the economy and some Treasury officials to this new body. It was given the privilege no other forecaster has of getting prior access to budget measures so when the budget is published the OBR can publish a set of forecasts that include the impact of the latest budget measures. Other independent forecasters can then catch up, putting the new budget measures into their models and running them to see what change results. The OBR forecasts and the average of private sector forecasts are often quite close to each other, with the old Treasury model still having some sway with a range of external economists. Treasury officials clearly work closely with OBR ones, as they used to do when they were all part of the same organisation.
The main problem with this system is the failure of the POBR to come up with reliable and accurate forecasts of the budget deficit. This matters hugely because their wrong forecasts unlike other people’s are used to mark the government’s homework. The main economic policy control is a derivative of the old Maastricht debt and deficit controls. The government aims to have debt falling as a proportion of GDP by the end of the five year forecast period if not before. This relies on the OBR forecasts of the difference between two large numbers, total spending and total revenue, five years hence.
In recent years the OBR has been £100bn or more out in its same year forecasts, let alone in its five year forecasts. The OBR presumes to say the government needs to raise an extra £10 to say £30bn in taxes, when its deficit forecasts swing by far more than these sums year by year. Observing the pattern they tend to greatly exaggerate the deficit when the economy is growing and underestimate it when the economy is slowing or shrinking. The main errors occur on the revenue side. Their model does not seem to take much account of the behavioural effects of higher tax rates which may depress tax take, or the way in which lower tax rates may boost tax take. It certainly doe snot seem to recognise the great sensitivity of revenues to gr9wth rates.
The independent OBR should follow what the Bank has decided. faced with its own failure to forecast inflation, crucial to its task, the Bank has announced a review of its models. The OBR needs to do the same, as it models cannot forecast deficits sensibly, leaving no sound basis for their advice on tax levels.