Bad lags and rent rants

FT Alphaville has repeatedly kicked up a fuss about how US shelter prices are captured in America’s CPI and PCE inflation measures. See here, here, and here. We can’t help it.

After all, rents and owner equivalent rents — a dubiously imputed measure designed to capture the cost of shelter for owner-occupiers, calculated using existing and new rents, phew! — are helping to keep US inflation above target, providing fodder for the hawks.

FTAV has also ranted about lags in various components of the inflation basket. Right now, America’s sticky CPI inflation holdouts include rents, OER and car insurance.

The chart below shows how price pressures are working their way through the car insurance ‘supply chain’.

The point? Can interest rates really make any difference to lagged costs just passing down the “supply-chain”? MainFT covered this in May.

Rental (and hence OER) components are still high because they partly reflect the lack of construction during the pandemic era. (New rents are falling, but since many are on long-term contracts, overall rent costs remain high.) Low housing supply remains a problem. Dearer car insurance now is also a product of past cost pressures for vehicle components.

These are supply matters, which rates can do little about. High rates could even exacerbate housing shortages by disincentivising construction.

Research by the San Francisco Fed suggests supply-driven inflation is now a more significant contributor to inflation than demand.

So, we have a situation where interest rates stay higher to target components that interest rates have little impact on, thereby squeezing down harder than necessary on rate sensitive components. That seems like a less than ideal way to conduct monetary policy.

Indeed, the “keep rates high until something breaks” approach is common through historic cycles.

A provisional July 2024 paper, entitled How should monetary policy respond to housing inflation? and written by Javier Bianchi, Alisdair McKay and Neil Malhotra at the Federal Reserve Bank of Minneapolis, concurs:

Our findings suggest a reappraisal of the measure of inflation targeted by central banks. The standard analysis in the New Keynesian literature finds that monetary policy should place more weight on the inflation in sectors with stickier prices because these sectors have more potential for misallocation. From our perspective, because the market for housing services is largely supply-determined, changes in prices are disconnected from the consumption of housing and are therefore less relevant for welfare.

But the logic extends beyond shelter to other sticky or lagged components. The below is from a May 2022 paper in Econometrica by Jennifer La’O and Alireza Tahbaz-Salehi:

Optimal [monetary] policy stabilizes a price index with greater weights assigned to larger, stickier, and more upstream industries, as well as industries with less sticky upstream suppliers but stickier downstream customers. In a calibrated version of the model, we find that implementing the optimal policy can result in quantitatively meaningful welfare gains.

There are a few upshots:

— A rigid focus on a 2 per cent inflation target can be problematic, when the spread, direction and nature of price pressures matter as much as the total level. Central bankers need to be able to articulate that.

— At some point in every rate-tightening cycle, monetary policy can switch from being useful in tackling inflation to being potentially harmful for the economy. Other policy levers such as fiscal policy and supply-side measures (eg housebuilding) may then need to pick up the slack.