Exclusive interviews with leading policymakers that convey the true policy message that impacts markets.
Reporting on key macro data at the time of release.
- Emerging MarketsEmerging Markets
Real-time insight of emerging markets in CEMEA, Asia and LatAm region
- MNI ResearchMNI Research
Actionable insight on monetary policy, balance sheet and inflation with focus on global issuance. Analysis on key political risk impacting the global markets.
- About Us
Real-time Actionable Insight
Get the latest on Central Bank Policy and FX & FI Markets to help inform both your strategic and tactical decision-making.Free Access
The U.S. Bureau of Labor Statistics could modify its calculation of CPI to correct an over-reliance on apartment rental data which has tended to push owners' equivalent rent slightly higher since 2013, Cleveland Fed economist Randal Verbrugge told MNI.
OER, the implicit cost to rent the home you own, will likely be revised down because apartment rental prices have accelerated much quicker than those of detached units, which account for only about 30% of the BLS' rental sample, he said. The BLS does not distinguish between the two types of housing, using only location to calculate neighborhood rent increases.
"We've discovered, we think for the first time ever, evidence that there could be a problem with just using 'location, location, location' as the basis for determining how much rent for a particular owner went up," he said, referring to research produced jointly with staff at the BLS.
Even after controlling for location, apartment rents in each neighborhood in the BLS's sample grew much faster than detached unit rents from 2013 to 2016, leading the agency to overestimate OER by about 0.3 pp per year, Verbrugge said in an interview last week. Between 2017 and 2020, the BLS overshot OER by roughly 0.2 pp to 0.25 pp per year, he added, though those calculations are less accurate.
That translates into an overestimation of the entire CPI by approximately 0.08% from 2013 to 2016 and 0.05% over the following three years, he said. Shelter prices account for roughly 30% of the total CPI basket.
While those may sound like almost negligible percentages, the CPI is utilized by critical programs like Social Security, and the BLS "will not tolerate that level of inaccuracy," Verbrugge said.
Verbrugge said he expects the BLS to address this issue in the near future, though the agency hasn't yet formally responded to the research. BLS did not respond to a request for comment.
A WORSENING SAMPLE
The BLS's rental sample could become even less representative following the pandemic, which drove many Americans out of crowded cities into more spacious suburbs that tend to have less apartment buildings and more single-family homes.
"It's possible that this representivity problem is being exacerbated by the current pandemic," Verbrugge said, and owners' equivalent rent may have even been underestimated over the past year because of insufficient aggregation weight on the suburbs and excessive weight on center cities.
"We can easily imagine that detached unit rent growth over the last year perhaps was much higher than apartment rent growth," he said. "But because you're putting most of the weight on the apartments, you'd miss that."
Owners' equivalent rent rose 0.3% in June, the same increase as May, and was up 2.9% from a year earlier. Rent of primary residence rose 0.2% over May and 1.9% from one year ago.
Sign up now for free access to this content.
Please enter your details below and select your areas of interest.