Recent Updates

  • US: Challenger Employment Report (Nov)
  • Turkey: ES Prices, Central Bank Balance Sheet, Debit and Credit Card Spending (Nov)
  • UK: Nationwide HPI, BoE Decision Maker Panel Survey (Nov)
  • Markit: Mfg PMIs: Ireland, China, India, Netherlands, Turkey, Poland,
  • more updates...

Economy in Brief

State Coincident Indicators
by Charles Steindel  April 3, 2018

On April 3 the Federal Reserve Bank of Philadelphia issued revised estimates of state coincident indicators, along with January values. The Philly Fed produces figures for all 50 states, as well as the nation as a whole, using a consistent methodology. The coincident index for a state or the U.S. as a whole is computed as the underlying factor that best summarizes the information in four variables: payroll employment, manufacturing hours, the unemployment rate, and real wages (using the national CPI to deflate). The indexes’ longer-term trends are equated to that of real GDP (state or national, depending on which is being computed). While there clearly could be a much larger body of information useful for summarizing the state of the economy of a state or the nation (the New York Fed’s comparable indexes for New York State, New Jersey, and New York City use different inputs for New York City), this methodology allows for a fairly straightforward comparison across the states.

The revised data incorporates the recent benchmark revisions in the labor market and wage data. The base year used for reporting the levels of the indexes was moved forward from 1992 to 2007; this change allows for a clear comparison of current levels to those around the past business cycle peak. Philadelphia estimates that national activity in January was 22% above its 2007 average (real GDP in the fourth quarter of 2017 is reported to have been 16% above the 2007 level). Three states—West Virginia, Louisiana, and Alaska—had levels of activity in January 2018 less than 10% higher than their 2007 level. On the up side, the Philadelphia Fed reports that the economies of Utah, Arizona, and New Hampshire have each grown more than 40% since 2007. Rather strikingly, the two largest state in population—California and Texas—are not far behind (Texas ranks fourth, California fifth). Some major traditional manufacturing states (Pennsylvania, Ohio, Indiana, Illinois, and Michigan), while not at the bottom of the list, are reported to have grown less than the nation over the last decade. Interestingly, so has New York, in contrast to the many stories about boom times in the Big Apple. Arguably, economic weakness upstate has been sufficient to make a real difference to the growth record for the Empire State as a whole.

As to the figures for growth over the last year (January 2017-January 2018), Louisiana and Alaska are still at the bottom (the point figure shows an outright contraction in activity in Alaska), while West Virginia remains near the bottom, at number 46. At the top of the heap are Alabama and Tennessee, whose indexes are reported to have grown more than 6%, compared to a nation-wide increase of 2.8%. The five states noted for unusually strong growth since 2007 (Utah, Arizona, New Hampshire, California, and Texas) also clocked increases since January 2017 higher than the nation as a whole. None of the industrial states mentioned as having soft growth over the last ten years showed gains notably higher than the nation in the last twelve months (nor did New York). Thus, the last year has not seen any marked change in growth patterns across the nation.

The Philly Fed also compiles indexes of leading indicators for the state, designed to predict growth in the coincident indexes. These have not yet been revised to reflect the new coincident indexes (the release is scheduled for April 5); the data to compile leading indexes at a state level are scarcer and arguable less reliable than those used for the construction of the coincident measures. National data, such as interest rate term premia, are used along with state information such as housing permits and unemployment claims, but such usage is dependent on the assumption of a reasonably reliable hig-frequency relationship between the national series and state activity.

large image