No need for negative SNB rates

A second approach is to ask how the Fed has behaved, on average, in similar macroeconomic circumstances in the recent past. This can be formalised by estimating a simple reaction function, a statistical model that links the Fed’s interest rate decisions to key economic variables. Such models provide a useful benchmark, but they cannot be used as a substitute for a careful assessment of the economic and financial environment facing policymakers.

We estimate such a model using data from every FOMC meeting since March 2022, a period that spans the peak of the inflation cycle, the subsequent tightening phase, and the initial easing last year. The model identifies two variables that have been empirically important for Fed decisions over this period.

The first is the real policy rate, defined as the nominal policy rate minus core inflation. This captures the overall restrictiveness of monetary policy relative to underlying price pressures. The second is initial jobless claims, which provide a timely signal of labour market conditions. Together, these variables summarise the trade off at the heart of the Fed’s dual mandate.

An additional benefit of this framework is that it allows an estimate of the neutral real interest rate, r*, defined here as the level of the real policy rate at which interest rates are just as likely to be cut as raised. Based on post 2022 data, we estimate real r* at around 0.5%. This implies a nominal neutral rate somewhere between 2.5% and 3%, depending on the inflation measure considered.

Applied to current data, the reaction function assigns an 81% probability to no change in interest rates at the January meeting and a 17% probability to a 25 basis point cut (see Chart 1). While slightly more dovish than market pricing, this assessment is broadly consistent with the CME probabilities and reinforces the expectation of a pause.

The key point is not the precise probabilities, but the message they convey. Both markets and the Fed’s own recent behaviour suggest that policy is close to neutral, that labour market conditions are not yet weak enough to force the Fed’s hand, and that inflation remains an important constraint. Absent a material deterioration in employment data or a sharper decline in inflation, the January meeting is likely to be uneventful.