When do central bank losses matter?

This week, an FT headline caught my eye: Treasury bails out BoE for first losses on QE programme 

It may ring some alarm bells, but I think it’s crucial to break down what is going on behind the headline and understand whether “losses” at the BoE should be a cause for concern. 

Pre and post QE

It’s important to remember that central banks have both balance sheets and income statements. Most of us know about the first, but we can sometimes forget about the second. 

Pre-QE, the two sides of a central bank balance sheet were simple. Liabilities were currency (carrying zero interest) and assets tended to be short-dated T-bills. Therefore, the balance sheet was small and the income statement was negligible. But once QE kicked in, central banks created vast amounts of money by buying long-term bonds with the cash they had created. 

The bonds sat on their balance sheet, paying interest to the central bank. The cash created ended up with private sector banks, who parked some of it (reserves) in an account back at the central bank. For the most part over the past decade, the central bank has paid no interest on reserves. In some cases, they even paid “negative” interest, charging banks to hold their money.  

Thus, the central bank was earning interest on its long-term holdings whilst paying nothing on its deposits. This turned a profit and, over the past decade, in both the US and the UK, these were paid back to the government Treasury, providing a cushion for the government’s finances.

A new and very different world 

Fast forward to 2022, post-COVID and with rising inflation, the world is very, very different. 

As we know, the BoE and Fed are both aggressively raising short-term interest rates, meaning they are paying out large interest payments on the reserves sitting on their balance sheet. At the same time, the bonds they bought in 2020 and 2021 during COVID were all super low-coupon bonds, so they don’t pay much interest at all. 

As a consequence, central banks are now beginning to lose money on their income statement, and the Treasury is having to start paying the money back to the central bank. I know this has happened in the UK, hence the headline in the FT. But is it any cause for concern? 

A problem or mere semantics? 

Some argue that the rearrangement is mere semantics or an accountancy foible. It’s the same government, so it doesn’t matter whether the central bank or the Treasury loses money. 

Prof William English and Donald Kohn argued this in a blog for Brookings this year, stating how the policy outcomes of QE tend to be more beneficial (and, often, necessary) than the prospect of a few quarters of losses at the central bank. Hence, there’s little to worry about, as most central banks have a buffer to absorb losses. Even if the losses get too great they can create more cash, so they contend that there isn’t really a problem. 

However, the cash will eventually end up at the central bank in the form of reserves, and the central bank needs to pay interest on it. And if an inverted yield curve persists, the losses could, in theory, outpace reserve creation, and the central bank’s capital buffer would be destroyed. By plugging the hole, the Treasury is essentially shortening the duration of the Treasury’s debt. 

We’ve written how the UK has the longest duration debt out of all developed markets at nearly 14 years. If HMTreasury is bailing out the BoE because of the mismatch between short and long rates, the Treasury’s liabilities will, effectively, shorten from 14 years to 2, according to the OBR. As a result, the government’s balance sheet becomes a whole lot more sensitive to rate rises.  

A political problem 

There are other examples of smaller central banks that lost money due to yield curve inversion and have, essentially, run a “negative equity” position as they wait for profits to return to rebuild capital. This paper from the US FED presents three case studies, highlighting the Swiss, Czech, and Chilean central banks, all of whom went through this situation in the past 30 years.

For me, the paper’s unwritten conclusion seems to be, considering how profits will eventually return, losses don’t matter as long as the political climate remains supportive of the role of the central bank. However, it remains to be seen if such a sanguine approach will apply to either the BoE or the FED, both central banks that are much more in the limelight due to their size and market influence. Both have also been used as a ‘political football’ by some in recent times. 

Certainly in the UK, the pressure for the BoE to remain independent and respected is coming under strain, most notably in political circles. Those who question its role should be careful, as the fire they’re playing with burns hotter in a post-QE world than, perhaps, they realise.