Paying for QE

In my article last week on LDI I predicted that the turmoil in the pension fund sector and the gilt market two months ago would not be the last uncomfortable moment for the authorities. And it already looks as though we can see where the next challenge will come from, as the question of the cost of QE, and in particular the amount of public money the Bank of England will need in the coming years to keep its balance sheet whole, is coming sharply to the fore.

This is a complex area, and it is probably fair to say that few outside the central banking world fully understand the mechanisms, or have even given the matter much thought to date. But the numbers are so large – on one estimate, the Treasury may need to provide over £130 billion to the Bank over the next 6 years or so – that I suspect the national media will in due course pick up on the story. And when they do, it has all the makings of a political nightmare for the government.

In order to understand what is going on, it is necessary to delve into the Bank’s balance sheet, and explore how the purchases of gilts and other assets has been managed under QE. And to do this we need to meet the Asset Purchase Facility.

The Asset Purchase Facility

Although QE was only introduced about a dozen years ago, the idea of the Bank buying market assets is far from new. Central banks have always bought and sold assets as an integral part of their interaction with markets, an activity that goes under the name “Open Market Operations” (OMOs). Classical OMOs are a tool of day to day market management not long term economic management though, and as such are usually fairly small, and usually of fairly short tenure – and traditionally they have been conducted through the central bank’s own balance sheet.

But when the Bank of England first contemplated QE operations in 2009, in response to the Global Financial Crisis (and in partnership with other central banks), it quickly realised that they could grow to be (a) very large and (b) outstanding for a long time. As a result it felt the need to separate these asset purchases from their more traditional OMOs, and it decided to conduct them through a new facility (the “Asset Purchase Facility” or APF), financed through a new wholly owned subsidiary of the Bank (“Bank of England Asset Purchase Facility Fund Limited” or BEAPFF).

(This was not only to stop them complicating and confusing the operation of the rest of the Bank’s balance sheet, but also because it made it easier to identify the profits and losses arising from QE, which was necessary because right from the start, the Bank secured an agreement with the Treasury that these profits and losses would be for the Treasury’s account – in other words, the Treasury would take any profits and bear any losses, thus making the Treasury the “true” owner of the securities acquired under QE in economic terms).

The mechanics of the asset purchases were then relatively straightforward. The Bank lent money to BEAPFF, which bought the assets from the market. BEAPFF thus had a simple balance sheet which was long the assets bought in the market, financed by a loan from the Bank. And crucially BEAPFF also had a guarantee from the Treasury that it would keep this balance sheet whole – a non-tangible but far from negligible asset. So BEAPFF was square.

The effect on the Bank’s balance sheet was almost as simple. It had an asset (the loan to BEAPFF), and in practice a matching liability in the form of “bankers’ balances”, the reserves the commercial banks leave with it. Both asset and liability were interest bearing at the Bank’s lending rate, and the Bank’s position was therefore also square.

Accountants among my readers will spot that in orthodox accounting, the Bank’s creation of BEAPFF and operation of the APF through it would not normally have made its balance sheet immune to the value of the APF; since BEAPFF is a wholly owned subsidiary, its accounts should normally be consolidated with the Bank’s. But the arrangement with the Treasury makes this unnecessary. In addition, if there was any doubt over the ability of BEAPFF to service its loan from the Bank, the Bank should have marked it as impaired and again, taken the consequences on its own balance sheet. But again, the Treasury guarantee enables the Bank to treat the loan as good regardless of BEAPFF’s financial position.

So in every sense the Bank has successfully isolated its own balance sheet from the consequences of QE and the resulting financial performance of BEAPFF. It is worth noting that the Bank (cognisant perhaps of its unusually low capital base compared to its peers) is almost alone amongst central banks in having agreed this position with the Treasury: most central banks were not so alert or quick to tie their finance ministries into their QE operations and many face considerable losses – in the Eurozone in particular, where finance ministries find it more difficult to recapitalise their central banks due to the rules of EMU, this may yet cause real problems.

The consequence of rising interest rates

For the first decade of QE, the operation was very profitable for BEAPFF, and so the Treasury. Using a loan at 0.1% to buy gilts yielding north of 1% produced large profits, and over the decade BEAPFF transferred around £120 billion in all to the Treasury, where it was both very welcome and entirely unremarked upon. No Chancellor felt the need to explain where this nice little revenue stream for him was coming from!

But times have changed, and with the rise in interest rates from 0.1% to 3.0%, BEAPFF’s financial position is looking much less healthy, and the value of the Bank’s arrangement with the Treasury, and the importance of the Treasury guarantee, has come into sharper relief.

As interest rates have risen, BEAPFF’s finances have suffered in two ways: on the asset side the bonds it has bought for the APF have fallen in value, and on the liability side the cost of servicing the loan from the Bank has risen sharply. The net result is a twofold hit: on a mark-to-market basis its balance sheet shows a large deficit (ie liabilities exceed assets), and its P&L account shows a running loss as the income from the assets is not enough to pay the interest due to the Bank in servicing the loan.

It is true that the mark-to-market loss will not materialise until the assets are either sold or mature. But the loss, as all investors should know, is no less real for not yet being crystallised, and it is the combination of both of these that leads to the calculations that the Treasury may need to find as much as £130 billion over the rest of the 2020s to keep BEAPFF solvent, and remove any knock-on effect from impacting the Bank.

If that was the end of the matter, the story would probably remain one of niche interest only, despite the huge sums involved. Transfers between Treasury and Bank are difficult for the man in the street (or therefore the tabloid press) to get very excited about, and questions of the Bank’s financial health, and what it even means if a central bank has a net negative balance sheet, are guaranteed to have eyes glazing over even in the most learned of financial circles. Several central banks have at various times in the past run net negative balance sheets, often for years, and it has not impacted their effectiveness, so the issue of central bank solvency and whether negative net worth matters for them is far from cut and dried even for specialists in the field.

The position of the bankers’ balances

But unfortunately it may not be the end of the matter, because one reason the Bank charges a market rate of interest on its loan to BEAPFF is that the counter-liabilities on its own balance sheet, those bankers’ balances, are also interest bearing. And this gives any mischievous journalist an easy target to attack: with interest rates at 3%, the interest paid to commercial banks on their balances at the Bank is now over £25 billion a year. Cue potential headlines about “subsidies to fat cat bankers” and “£2 billion a month of public money for doing nothing while they pay their depositors peanuts”.

If the Bank did not pay this interest it could be argued that it would not have to levy interest on the loan to BEAPFF, BEAPFF would not be racking up the losses it is and the Treasury would be off the hook.

It is, as I say, an easy line of attack. And what makes such an attack even more plausible and attractive for any journalist seeking to cause trouble is that until recently central banks did not pay interest on bankers’ balances. The Bank of England only started doing so in 2006, and the Federal Reserve (who have similar issues) even more recently in 2011. Why can we not, people ask, go back to the position 15 years ago or so when these balances were not remunerated?  Problem then solved?

Alas no. As Huw Pill, the Bank’s chief economist, explained at length on Thursday last week, the move to paying interest on bankers’ balances, which in the UK at least predates any consideration of QE, is more to do with maintaining control of short term interest rates and thus monetary policy. Pill’s comments were surprisingly defensive, playing heavily on the “contractual obligation” to honour the Bank’s agreement with the commercial banks (as if changing the interest rate broke that agreement), and I suspect he knows this is a sticky wicket for the Bank for all the presentational reasons I have set out above. But the essence of his remarks is correct.

This is because if the Bank stopped paying interest on the reserves the commercial banks would be incentivised to use their enormous reserves (far more than they need for balance sheet management or payment system purposes) to buy short-dated assets (treasury bills, short gilts), which would reduce the yield on them very sharply. This would hugely complicate the Bank’s control of short interest rates and render monetary policy very much more complex, if not wholly inoperative.

And so (exactly as in my article last week) the chain of unintended consequences grows. In order to make the banks safer their permitted leverage ratios have been tightly circumscribed, making loans to customers less attractive for them and reserves at the central bank more attractive. These therefore grew, and in order to retain control of monetary policy the Bank agreed to remunerate them. Then QE caused them to grow several orders of magnitude larger, and the final result is that what looked in 2006 like a small technical adjustment when bankers’ balances were a small fraction of their current £830 billion, has turned into today’s £25 billion a year in interest payments.

Politics and economics

In summary, well-meant technical adjustments to the banking regime (paying interest on bankers’ balances) and the expedient of QE (rescuing the economy after the Global Financial Crisis), while both eminently sensible and defensible at the time, are coming back to bite the politicians many years later.

It does not matter that for the first dozen years of QE, the operation was very profitable for the authorities. As I noted above, BEAPFF has transferred about £120 billion to the Treasury in profits since 2010, thus making their calls on the Exchequer now more a matter of timing of flows in and out than any wholly new money. Nor does the argument, so valiantly extended by the Bank last week, that paying interest on bankers’ balances is a technical matter and far better than not paying it, cut much ice either. And finally the fact that in other countries, central banks and their finance ministries are if anything facing even more serious problems and have handled the matter less adeptly is of course irrelevant for critics of the government. Such arguments are entirely valid in economics and entirely without traction in the febrile world of politics.

All the public, and any tabloid journalist out to stir up trouble, will see is that in a time of general austerity and higher taxes for all, the authorities are paying a large amount to the commercial banks “for doing nothing” and an even larger amount to the Bank of England “to help it overcome the consequences of its own higher interest rates”. Would that mortgage payers were so lucky!

At a time when he is pursuing unpopular austerity policies and struggling to keep both the country and his parliamentary colleagues onside, poor Jeremy Hunt needs headlines like that like a hole in the head.

 

I am grateful for help and advice with this article from Robin Darbyshire, a former Bank of England colleague and now an independent consultant on central banking finance and accounting issues