Frank Van Lerven is a programme lead of macroeconomics and Dominic Caddick is an assistant researcher at The New Economics Foundation, a British think-tank. Their post is in response to an FT Alphaville article, Monetary Policy on the Cheap, that referenced NEF’s research The fiscal implications of interest rate risk from the Bank of England’s balance sheet expansion have repeatedly been touted as a key reason for future fiscal belt-tightening. Prior to the 2007-2008 financial crisis, the BOE did not pay interest on all the reserves held by banks — the current state is an exception, not the historic norm. While many fears of debt sustainability and interest rate rises are largely misplaced, paying interest on reserves is a policy choice, one that merits adequate debate and scrutiny.The government’s debt servicing costs are set to increase because quantitative easing is financed by central bank money creation that pays out the Bank’s policy rate of interest. With nearly £1tn deposited by commercial banks overnight as a result of QE, the banking sector is now set to receive £20bn by the end of 2023 and just under £100bn by the end of 2027.
These are not trivial sums, and the optics of boosting the banking sector’s profits while families across the UK are struggling to make ends meet is eyebrow-raising to say the least. But there is also an inherent contradiction: with one hand the government has referenced exposure to the Bank’s rising interest bill to justify fiscal tightening, while with the other it is making billions of income transfers, most likely directly passed onto banks’ bottom-line earnings, for arguably no additional services rendered.Central bank reserves are often misleadingly referred to as a form of government ‘debt’, with the implication that banks deserve interest compensation for holding these liabilities. But, unlike gilts, no principal was ever borrowed. Central bank reserves are perpetual, in that they never mature (nothing has to be paid back), and banks face no credit risk. In addition, while a number of macroeconomic forces determine the interest paid on gilts, the rate of interest paid on reserves is determined solely by the central bank.Reserve remuneration therefore is not the result of a commercial bank providing any material services to the Bank, and the payment of interest is not the consequence of any financial obligation, like paying back a debt. The income transfers to the banking sector and the ballooning interest bill accrued by the Bank are therefore a policy choice, based on the desire to influence money markets and credit conditions. Paying interest on reserves acted as a floor on market rates in an environment of ample reserves, allowing the Bank to set the interest rate and change the amount of central bank reserves available to the banking system independently of one another.While the Bank has already made plans to unwind its QE programme, one way to avoid such considerable income transfers to banks would be a quick sell-off of its current bond holdings. But this option seems unfeasible, not least because it could jeopardise monetary and financial stability. It would also be unnecessarily expensive, substantially increasing the net interest servicing costs of the government, and would result in the Bank making significant losses (roughly between £100-200bn) that would have to be covered by the Treasury.In a recent report for the New Economics Foundation, we argue that a better option, grounded in the recent innovations in reserve management by the Bank of Japan and the European Central Bank, would be to implement a ‘tiered reserve system’. Specifically, the Bank could only pay interest on a portion of central bank reserves, or it could stop paying interest altogether.This would permit the distinct separation of the Bank’s policy rate (allowing it to retain control over money market interest rates) from the government’s interest servicing costs and the profitability of the banking sector.Transitioning to a tiered reserves framework has its own costs and benefits. A slightly un-nuanced restatement of these trade-offs was recently made by Toby Nangle and Tony Yates. Most notably, moving towards a tiered reserves framework could entail an implicit “tax on credit intermediation”. Once the Bank stops remunerating interest on a certain portion or (or all) reserves then, when market rates move significantly above zero, commercial banks would have interest-bearing liabilities (customer deposits) but no interest bearing assets (central bank reserves) to cover the interest owed and costs of administering such deposits (especially those created via QE.Commercial banks could pass this cost onto savers through lower interest payments. However, they most likely would want to still attract savers to maintain market share and reduce exposure to deposit migration, which a narrow and undiversified customer base could exacerbate.Banks, thus, would most likely pass the increased cost of higher interest rates onto borrowers, just as they did in the past when they did not remunerate reserves. This is, however, the whole point of raising rates to begin with. For these reasons, we posit that a tiered reserve system can act as a possible automatic stabiliser for price stability. By amplifying the Bank’s anticipated contractionary effect of raising rates, this would be a “feature not a bug”, as aptly put in a recent IMF working paper.The withdrawal of such sizeable income transfers to the banking sector merits consideration, particularly given the many benefits the banking sector already enjoys. These include access to the central bank’s payment system, the dramatic reduction in banks’ funding costs through QE (ie, the cost of acquiring reserves), significant indirect subsidies from the Bank and wider government in the form of credit guarantees (lender of last resort function by the Bank) and liquidity guarantees (deposit insurance), and the ability to create money (bank deposits) via lending.More importantly, in the existing floor system, banks would raise interest rates for their borrowers anyhow, while hardly passing on rate rises to their creditors. However, they would still benefit from significant income transfers from the Bank, at the expense of the government and the taxpayer. So it’s likely that banks are the ones currently getting a free lunch.If we are unhappy with the Bank making billions of income transfers to the banking sector why not just tax the banks? A fair question. But, notwithstanding the current low appetite for higher taxes, surely it’s equally fair to question the design of the system to begin with, and address the issue at its source.Some worry that tiering reserves could amount to fiscal policy through the back door. However, millions of pounds in income transfers to the banking sector, for no services rendered, is already a form of fiscal policy, and one that is less aligned with the public good and societal interests.In any case, the choice to keep remunerating bank reserves while touting interest rate risks from QE as means to justify future fiscal tightening will be a reflection of priorities, not an economic necessity. More
