Dissecting the Bank of Japan’s ETF splurge

Receive free Bank of Japan updates

We’ll send you a myFT Daily Digest email rounding up the latest Bank of Japan news every morning.

The most unorthodox bit of the Bank of Japan’s unorthodox monetary policy is its purchases of Japanese equity ETFs, amassing a portfolio that booked at about JPY37tn ($260bn) at the end of July.

It’s far from the only weird legacy of a weird decade for monetary policy. The Swiss National Bank has amassed in $214bn global stocks, for example, while the Federal Reserve in March 2020 started buying some bond ETFs. And then there’s Turkey, which, well, keeps being Turkey.

But most serious central banks have been loath to touch anything but top-rated government debt. The Fed only bought about $8bn of bond ETFs at the depths of the Covid panic, and quickly ditched them. Because the SNB has primarily been interested in stemming the franc’s appreciation, it only purchased overseas equities, making it a small fish in a big sea.

The BoJ’s haul of Japanese equity ETFs is pretty wild, especially when you consider that the book value is actually below the ca JPY52tn of market value as of the end of March. That’s about $366bn in dollar terms, and equal to about 6 per cent of the entire Japanese stock market.

However, an interesting new paper from the BIS points out why this hasn’t had the market impact that you might assume. Spoiler alert: it’s because of securities lending. Here’s the money shot:

Using firm-level panel data, we find that the BOJ’s purchases raised stock returns more for those stocks with limited availability in the stock lending market. Nonetheless, over the longer term, the BOJ’s accumulated purchases lowered lending fees and weakened the effects of their purchases on stock returns. This result suggests that ETF managers supply stocks that constitute ETFs held by the BOJ to the stock lending market, which weakens the policy effects of the program.

(On an unrelated note, we want to tell those responsible for formatting BIS papers that there is a circle of hell reserved for people that don’t use spaces in digital reports, which makes copying and pasting chunks of texts a diabolical exercise).

Let’s unpick that a little. Basically, the paper argues that the BoJ’s actual ETF purchases do have an impact (as you’d expect from a price-insensitive whale buying a set amount every month), but the effect dissipates quickly because the ETFs then lend out the individual shares. Traders then borrow the overvalued shares and sell them short.

The net effect is to at least partly neuter the intended effect of the BoJ’s ETFs purchases, BIS economists Mitsuru Katagiri, Junnosuke Shino and Koji Takahashi argue.

Over longer periods, the BOJ’s accumulated purchases lowered lending fees in the stock lending market and weakened the effects of the BOJ’s purchases on stock returns, particularly for those stocks with limited availability in the stock lending market. This result suggests that the accumulated purchases by the BOJ increased the supply of lendable shares in the stock lending market, thus diminishing the policy effect of the ETF purchase program. We call the rightward shift of the supply curve in the lending market the “lending channel.” These results on the lending channel indicate that the intended effects of the BOJ’s ETF purchase program on stock returns were somewhat offset by the increased supply of lendable shares in the stock lending market through the response of ETF managers.

Huh. It’s almost as if short selling might play a useful role in keeping markets at least somewhat efficient?

Anyway, there is naturally some debate in Japan about what to do with all those ETFs, which don’t mature like bonds and therefore cannot just be rolled off quietly, raising the awkward possibility of the BoJ being a major long-term owner of a swath of Japan Inc.

This summer the finance minister floated an interesting trial ballon: perhaps the government could simply buy them off the BoJ at the more modest book value, and either flip them at current market prices for a quick gain (charitably assuming the price impact doesn’t kill the profits) or redistribute them to younger Japanese.

Which, you’d assume, would be politically attractive and fit well with Japan’s role in breaking new monetary ground.