Accounting objections to cash transfers don’t add up, and if CBs bought equity they could reduce inequality
There is growing recognition – most recently by Janet Yellen – that should another recession strike when global interest rates are this low, central banks will have to embark on either more aggressive forms of what they have already done directly and by stealth, or something new. Regarding the latter, some form of direct support for consumption, looks like the best candidate.
One of the main objections to cash transfers as a means to support household consumption is that it impairs the central bank’s balance sheet. It is important to put this pedantic objection to bed. On reflection, it is clear that the central bank balance sheet consequences of a base money financed cash transfer to households is in fact the same as (successful) quantitative easing (QE) via government bond purchases.
Here’s the background: One objection to central banks making cash transfers to households is that it creates an accounting liability (bank reserves) without a corresponding asset on the central bank’s balance sheet. In conventional accounting this implies a decline in equity. (I have argued that conventional accounting is nonsense when applied to central bank balance sheet, but let’s go along with convention to illustrate the point). So when a central bank makes a cash transfer to households accounting ‘liabilities’ rise, and assets are unchanged. By definition, accounting equity falls. What happens with QE? Despite appearances, the effects are the same – as long as the QE works. Successful QE involves a necessary loss on the government bonds purchased – something frequently overlooked. Why? If QE provides a successful stimulus to the economy, real interest rate structures should end up higher than prior to the stimulus, after an initial decline in term premia. If this is not the case, there was further room for the central bank to reduce real interest rate structures by either cutting rates or committing to keep them lower. It is reasonable to expect government bond yields to decline on the announcement of QE – if it functions either a signal for the future path of rates, or by depressing term premia. But if this stimulus is successful, yields should rise after the bonds are purchased and the economy recovers. The Bank of England’s post Brexit referendum QE is the classic example of what I am describing. Successful QE, if reflected in higher than previously expected growth, should result in higher than expected real interest rate structures at least for the market’s proxy of the risk free rate. The BoE has indeed made capital losses on its post-Brexit gilt purchases.
I am assuming that long duration government bonds are bought under the QE programme, and that the intention is to stimulate the economy through lower long-term real interest rates. This is not QE aimed at bringing down artificially high money market rates caused by a shortage of base money. There are more caveats. The central bank can be ‘lucky’ (or unlucky – depending on your perspective) if there is an unrelated shock to equilibrium real interest rate structures which reduces reduces yields. In this circumstance, they could fortuitously make a capital gain. This has happened in many instances since the financial crisis. These capital gains are accidents, not design.
Some will argue the losses I am describing are ‘only’ mark-to-market losses. This is irrelevant. If QE is successful and needs to be reversed, a mark-to-market loss become a realised loss. So the balance sheet loss of equity is identical to that of a cash transfer, of a similar magnitude.
Those who object to cash transfers on the grounds that they unique impair the central bank’s balance sheet, are barking up the wrong tree. The only significant difference in the balance effects of cash transfers compared to QE, is that the losses in the case of QE are contingent.
QE appears as a somewhat cack-handed and incompetent alternative to cash transfers. Another option, of course, as Roger Farmer points out, is that the central bank buys equities to stimulate demand when recession threat raises its head, rather than bonds. There is much wisdom to this. Not least if the policy works, the state will make a large capital gain, because reduction in the threat of recessions doesn’t just cause real rates to rise it also causes higher stock prices. And as Mark Blyth and I have argued, this capital gain could easily be returned to the lowest deciles in the distribution of wealth – who often bear the brunt of recession.
Having written this short note, I subsequently discovered a very similar argument has already been made by Simon Wren-Lewis. I shouldn’t be surprised. Simon focuses on the distributional differences.