A recent article (N. Brophy, Appleton Post-Crescent) outlined some of the causes and implications of heightened inflation. The article lays out some of Wisconsin-specific effects. The discussion is somewhat constrained since BLS only reports limited region-specific CPI data, and none limited to Wisconsin, so the author makes some inferences linked to housing prices, energy and wage costs. Nonetheless, there are some interesting regional differences.
Here is a comparison of rates, with a zoom into the closest to Wisconsin (East North Central division of the Midwest region (breakdown here). First, year-on-year, second month-on-month annualized, through November.
Figure 1: CPI-all inflation, year-on-year for nation (bold black), Northeast region (orange), Midwest region (green), East North Central division of Midwest region (red), all in % (decimal format), using not seasonally adjusted data. NBER defined recession dates peak to trough shaded gray. Source: BLS, NBER and author’s calculations.
As noted earlier, year-on-year inflation rates can obscure more recent developments.
Figure 2: CPI-all inflation, month-on-month annualized for nation (bold black), Northeast region (orange), Midwest region (green), East North Central division of Midwest region (red), all in % (decimal format), using seasonally adjusted data. Seasonal adjustment using Census X12 by author, except for national CPI. NBER defined recession dates peak to trough shaded gray. Source: BLS, NBER and author’s calculations.
The cost of housing continues to increase as well. In Wisconsin, the median home price is up 7.6% from November 2020, according to the Wisconsin Realtor’s Association.
Natural gas is up about 25% over the last year, and electricity costs are 5% higher.
Here’s a map of inflation differentials, by BLS divisions.
Figure 3: November CPI inflation rates by BLS division, year-on-year, %. Source: BLS.
Is there a correlation of inflation rates with observable factors? One candidate variable noted is vaccination rates. Using end-June partial vaccination rates, one finds a negative correlation – higher inflation is associated with lower vaccination rates. A quantile regression indicates each one percentage point increase in partial vaccination rates is associated with a 0.5 0.05% [corrected 12/28, h/t AS] percentage point decrease in y/y inflation, with a t-statistics of 6.0 (simple OLS yields a similar estimate).
See also Wisconsin experts, in E. Gunn (Wisconsin Examiner) on inflation.
Update, 12/27 10:30am Pacific:
Reader Econned asks for details off the regression results.
For the simple bivariate quantile regression results reported above.
If augmented with some control variables used in some cross-state growth regressions (see this post), one obtains slightly higher adjusted R-squared, and slightly smaller coefficients on the vaccine rate (as of end-June).
Where the control variables are:
mfg is ratio of manufacturing output to GDP in 2019, from BEA state level GDP data)
ldensity Log population density
wet Precipitation (less precipitation = higher values)
mild Temperature extremes (less extreme = higher values)
distance Proximity to water (closer = higher values)
By the way, if I estimate using two-stage least squares the bivariate regression specification, but instrumenting using mfg, ldensity, wet, mild and distance as IV’s, I still obtain a significant negative coefficient on vaccination rates.
(See this post for description of ldensity, wet mild, distance data provided by Professor Neumark, and file of control variables).
Update, 10:45am Pacific:
If you are wondering if states that voted for Trump in the 2020 election are experiencing higher inflation, then the answer would be yes.
If you are wondering if vaccination rates should be treated as endogenous, the answer is yes, although the rub is in selecting an appropriate instrument. I use a dummy variable, whether the state fought on the side of the Confederacy (so that one could arguably say that variable was predetermined with respect to vaccination rates at least).