Risk of window dressing: quarter-end spikes in the Japanese yen Libor-OIS spread

Abstract

It is well documented in literature that funding conditions have been subject to the undue influence of distorted incentives of banks to lend and borrow at quarter ends under the Basel III regulatory framework. We investigate whether or not funding risk possibly also suffers the same. Using a state space model, we find quarter-end spikes in the Japanese yen Libor-OIS spread, which arguably reflect a higher funding risk premium at quarter ends, during the global financial crisis and in recent years. The phenomenon in the former episode suggests that quarter-end reporting under Basel II might already have had an effect on the functioning of funding markets, as banks found the capital ratio requirement sharply more binding or constraining. The spikes in the latter episode, which are attributable to the effect of the leverage ratio requirement under Basel III, are found to be negative, partly reflecting the scarcity of high-quality collaterals against the backdrop of a large-scale asset-purchase programme introduced by the Bank of Japan and partly a negative interest rate environment. The evidence adds to the argument in favour of supervisory practices that require banks to report/disclose their average leverage ratio for the quarter instead of their ratio for the last day of the quarter. Indeed, despite the currently proposed reform, the capital ratio remains quarter-end-based, so there could still be quarter-end spikes in funding risk premium in times of financial adversity.