Housing and Core Inflation – It’s Complicated

Housing and Core Inflation – It’s Complicated

Laura Rosner - Market Economics
US Daily Spotlight | 22 Feb 2013 03:15 |

Markets were risk-off on Thursday, as equity markets sold off by more than 0.6% for the second day in a row, while the 10yr Treasury closed the day at 1.98%. Data for the day were unimpressive with jobless claims for the week suggesting only modest improvement in February payroll employment, existing home sales edging marginally higher and the February Philly Fed manufacturing index falling sharply (details of the report were less negative than the headline index implied). Meanwhile, core inflation was slightly stronger than expected in the January CPI report, mainly reflecting a rebound in core goods prices, but also modestly firmer increases in housing costs.

Two CPI indices, owners’ equivalent rent of primary residence (OER) and rent of primary residence (rent), measure changes in housing or shelter costs. Both components are estimated based on a sample of residential rents the BLS collects periodically. Owners’ equivalent rent is meant to capture ”the implicit rent that owner occupants would have to pay if they were renting their homes.” Even though the two indexes are estimated from rental market data, housing trends can, and often do, influence shelter cost trends in the CPI. This is because there is often a close relationship between the renter-occupied and owner-occupied housing markets. The implications of continued improvement in housing for rental housing demand, and thus shelter costs in the CPI, are important to consider, as the two CPI measures of rents make up more than 40% of the core CPI.

The relationship between house prices and owner’s equivalent rent inflation can be unstable, but strength in housing typically implies weaker rental demand. However, in this a-typical, investor-driven housing cycle, is unclear what the implications of a continued recovery in housing are for OER or whether, and to what extent, our housing forecast poses downside risk to our forecast for inflation.

Looking across different regions, we are seeing regional OER trends running in line with regional trends in rentership and rental housing supply. Rentership rates, which are one proxy for rental housing demand, have increased in all areas, and we think there is a good chance this shift will be sustained over the medium term. Relative to the rest of the country, the West and the South have seen the largest increases in rentership since 2006 Q4. These areas have experienced more extreme housing booms and busts, and so we would expect to see a heavier inflow of renters in these regions. The Northeast has seen somewhat less of a shift from homeownership to rentership, probably because the region has a more stable population of renters vs other areas. At the national level, we think the rentership rate is likely to remain elevated for some time. However, even if the pendulum has swung too far from home-owning to renting and we get some correction, this is unlikely to happen rapidly. Homeownership and rentership rates are slow moving variables that do not shift overnight.

Rental vacancy rates, which are one rough indicator of rental housing supply, are either at low levels or still trending down and suggest there is not a glut of rental supply hitting the market. While we are seeing builders respond to increased rental demand by adding to the stock of multifamily rental housing, so far this increase has not eased any of the upward pressure on rents. Furthermore, data on January housing starts revealed that some of the recent strength in multifamily housing starts probably owed to a change in building codes, so the increases in supply we have seen to date may not be sustained. Our read of historical CPI data suggest that rental demand, not supply, is perhaps more important for OER trends. Furthermore, it is difficult to gauge the extent of excess rental housing supply or if this supply is even located in areas that are heavily weighted in the CPI.

Indicators of rental housing demand and supply at this point suggest rental market conditions may remain buoyant for some time. Thus, shelter costs in the CPI will likely continue to be an important buffer to sluggishness in some of the more cyclically sensitive price categories. Our current forecast is for shelter costs in the CPI to neither accelerate nor weaken. Our assessment is that the implications of stronger housing on the core CPI are neutral for the time being, though this will be an important trend to monitor going forward.