Untangling liquidity cycle asymmetry

The 2022-2024 rate hiking and cutting cycle is an excellent case study for monetary policy transmission. For one, it was the fastest series of interest rate increases since the establishment of the euro. These hikes were met with a similarly prolonged cutting phase that has brought rates to their current neutral territory.

I became interested in the concept of symmetry with respect to policy transmission: was there a difference in how rates behaved on the way up as opposed to down? The particular aspect I focus on is the effectiveness of transmission from key ECB interest rates to interbank rates, the fastest and most immediate reflection of monetary policy in the economy at large. This will be done through examining the DFR and €STR.

Monetary policy is steered through the deposit facility rate (DFR). This is the rate at which monetary financial institutions (MFIs) can leave excess liquidity in the ECB’s safe hands—accruing interest in the meantime. The euro short-term rate (€STR) is, in turn, a composite of the cost of interbank borrowing. Only approved Eurosystem financial institutions can access ECB standing facilities; if banks cannot (or due to stigma do not want to) borrow from the central bank, they can borrow and deposit with each other.

As shown by the graph above, the €STR follows closely below the DFR. This is because large MFIs that have access to ECB facilities offer the service of taking on other’s excess liquidity, which they then deposit in the DFR. In effect they accept deposits (or rather take loans, i.e. €STR) and make an arbitrage profit from the higher DFR rate. If the DFR and €STR were the same, there would be no incentive to facilitate deposits for smaller institutions.

The story gets interesting with the introduction of the spread between the DFR and €STR. At the peak of monetary tightness, their difference was at its highest. This coincides with a period of remarkable profitability for large institutions. My running assumption was that it is a reflection of a price setting ability of large banks, whereby rising DFR is met with a respectively slower increase in interbank lending rates. The higher the spread, the greater their profit after all. The opposite is true as liquidity conditions loosened up.

That begs the question: was there a difference, if any, in how the spread reacted on both sides of the peak? Did the €STR behave in a comparable manner?

As it turns out, calculating the rate of change of the spread with respect to the change in the DFR in both periods results in a slope that is almost twice as steep for the cut as opposed to the hike. In other words, the spread was aggressively more responsive on the way down the mountain. This in turn means that interbank rates were relatively stickier during the rate cutting phase, allowing the spread to close in more quickly. Tinkering with the maths shows that policy passthrough from DFR to €STR was an average of 99.7% effective during the hike and 99.5% for the cut.

The vast difference (ca. twice as steep) comes down to interesting theoretical explanations. On the way down, the €STR is limited by the lower effective bound—which becomes more important at lower rates. The increase in possibilities as liquidity conditions become laxer is exponential, with financial institutions being able to do far more with their collateral and alternative sources of liquidity. Thus, they are not tied to the €STR, which moves more slowly. On the way up, financial corporations face diminishing possibilities with each ECB basis point change. They become more bound to the €STR, leading to stronger passthrough.

Alternatively, one can posit that when rates are going down, the opportunity cost of large institutions holding deposits at DFR increases rapidly. Thus, as opportunities are scarce, financial corporations without access to the DFR have more power to lend at rates closer to the DFR. This causes a more aggressive convergence to a small spread. The dynamic is reversed in a hiking cycle.

Analysing liquidity asymmetry is both intellectually stimulating and policy-relevant. Understanding why and how market responses differ during monetary policy cycles helps policymakers better hone their tools. As has been shown, interbank rates were considerably more responsive (in relative terms) during the hiking phase of the last monetary policy rollercoaster; the idea that monetary policy is more effective in stifling demand rather than stimulating it is a widely known phenomenon.

In the future, this analysis could benefit from an assessment of the market structure of interbank lending in the eurozone, and its role in influencing €STR price-setting. In the meantime, we are left waiting for the next policy cycle to test whether these findings hold true.