The Bureau of Labor Statistics has completed its annual review and revision of Local Area Unemployment Statistics (LAUS) for 2012-2016. The data revisions incorporate updated population controls from the U.S. Census Bureau, revisions from other original data sources, and model re-estimation.
For Arkansas, the revisions were not surprising. The originally published data showed a sharp increase in household employment over the first three months of 2016 — a reading about which we were very skeptical from the outset. After revision, the unprecedented surge in employment and labor force was eliminated completely. The new data show more robust growth in employment over most of 2015 and a much smaller upswing/downswing during 2016. The new data also show that the number of unemployed was lower than previously reported for much of 2015, with a smaller decline in the number of unemployed during the first part of 2016.

Source: Bureau of Labor Statistics, Local Area Unemployment Statistics (LAUS)
As a result of the revisions, Arkansas’ unemployment rate now shows a more rapid decline during 2015, followed by a more stable rate during 2016. The original data had indicated that the Arkansas unemployment rate declined to as low as 3.8% in May 2016, drifting back up to 4.0% toward the end of the year. After revision, the unemployment rate changed little over the year, starting at 4.1% then dropping to 4.0% for most of the rest of the year.

Source: Bureau of Labor Statistics
Today’s data revisions also included the official annual averages for unemployment. The report noted at least three distinctions for Arkansas:
1. From 2015 to 2016 (annual averages), the Arkansas unemployment rate declined 1.1%, from 5.1% to 4.0%. Massachusetts and South Carolina were the only states to show a larger decline (-1.2%).
2. For the year, Arkansas unemployment rate of 4.0% can be considered statistically significantly lower than the U.S. average of 4.9%.
3. Arkansas was one of only 14 states to show a statistically significant increase in the employment-population ratio (up 0.7 percentage points, from 55.1% to 55.8%)