Since late 2012-13, excess cash held in the Bank of England’s Asset Purchase Facility (APF) has been transferred to the Exchequer on an ongoing basis.

Transfers up to the level of the Bank’s income in the previous year are treated as dividends, reducing net borrowing; beyond that, they are financial transactions, reducing the net cash requirement but not borrowing. Future payments made from the Exchequer to the APF will be treated as capital grants, increasing net borrowing but not affecting the current budget.

To estimate the size of future flows, we have to make assumptions about when quantitative easing (QE) will be unwound and how quickly. Our approach to this is unchanged since March. We assume QE remains at £375 billion and that it begins to be unwound once Bank Rate rises above 1 per cent, with sales evenly paced at £10 billion a quarter thereafter. We also assume redemptions will not be reinvested once sales begin. The first sale is now assumed to be in the fourth quarter of 2015, two quarters earlier than assumed in March.

Our projection for the overall net transfer to the Treasury is now £40 billion, down from our March projection of £45 billion. The change can largely be explained by higher gilt rates, which imply lower gilt prices at the point of sale and therefore greater capital losses.

There is huge uncertainty about the timing and pace of QE unwinding and our assumptions should be regarded as a neutral way of illustrating the potential fiscal impact of the QE, rather than as a forecast of how the Bank of England is likely to act. The estimates of the overall net transfer to the Exchequer are also highly sensitive to changes in gilt rates. In a scenario where gilt rates rise by 200 basis points when QE unwinding starts, the Treasury would receive £61 billion, only to pay back £58 billion over the following years, giving a net transfer of £4 billion.

Table A: Fiscal impact of projected APF flows

Table B: Changes to the fiscal impact of projected APF flows since March