Will-O'-the-Wisp: The economic effects of the US government closure

The shuttering of one quarter of the federal government from December until late January dominated news in the United States. Despite the extensive political coverage, economic analysis was largely conjecture. The statistics for the period had not yet been released.  

With the shutdown limited to a minority of government functions, the only concrete fact was that 420,000 federal workers did not receive their salary and another 380,000 were furloughed without pay. 

Would the political wrangling in Washington cause American consumers to lose heart and pull back on spending? Would businesses curtail their hiring, sending a pall over the economy?

Until recently we did not know the answer. Though some information is still pending, delayed by the absent government statisticians, we have enough to put the drama in economic perspective.

Shutdown Tracking: Hard versus soft numbers

The shutdown has coincided with significant declines in the so-called soft numbers, consumer and business sentiment. The purchasing managers’ index (PMI) from the Institute for Supply Management (ISM) for services and the Business Optimism Index  from the National Federation of Independent Business saw sharp declines in December as did the consumer sentiment indexes from the University of Michigan and the Conference Board.  The ISM manufacturing index fell in the second half of the year then recovered in January.

Hard numbers for the labor market, non-farm payrolls, ADP payrolls and jobless claims remained strong in December and January.  Annual average hourly earnings were at the top of their decade range.

The largest unknown is the consumer sector particularly in December, the all-important holiday season.  Consumer credit from the Federal Reserve was marginally lower in the month. Beyond that the statistics are pending.  Retail sales for December from the Census Bureau will be delivered on February 14^th, with January’s numbers somewhere from the 15^th to the 21^st. Durable goods for December, also from Census will be released between the 14^th to the 21^st.

Business Sentiment Indexes

Purchasing manager’s indexes from ISM are surveys that ask business managers about current conditions, outlook and future plans.  The do not query actual numbers but intentions and sentiment. The other sentiment indexes are formulated in a similar fashion

The manufacturing PMI index rose to 56.6 in January from 54.3 in December which had been the lowest score since December 2016.   Manufacturing had an exceptionally strong two and a half years from July 2017 until November 2018, with the best sustained readings since 2003 and 2004.  The 12-month moving average in August was the fourth highest in 35 years topped in that period only by the three readings late in 2004.

ISM Manufacturing PMI

Reuters

The decline in the manufacturing PMI index began in September well before the government shutdown with the fall to 59.5 from August’s 14 year top of  60.8 and continued until December. As it preceded the closure and has partially reversed in January it is unrelated to the actions in DC.  

The ISM non-manufacturing index exhibits a similar pattern. The index peaked at a 13 year high in September at 60.8, declined to 60.4 two months later and then had a precipitous drop to 56.7 in January equaling the previous July low.   

The sharp fall in December and January and the lack of recovery makes it a candidate for shutdown impact. Nonetheless it remains on the upper side the post-recession range and well above the 50 division between expansion and contraction.  The December 12-month moving average score of 58.82 is the highest in the series which dates back to July 1997.

ISM Service PMI

Reuters

The National Federation of Independent Business (NFIB) Business Optimism Index lines up with the above surveys.  It soared in late 2016 from 94.1 in September to 105.8 in December the steepest rise in the 44 year series. It continued higher with occasional setbacks peaking at 108.8 in August 2018 an all-time record for this statistic.

The January score of 104.4 brings it back to the range of the initial burst in 2016. It remains above all post-recession readings prior to that point and also the vast majority of scores in the series history. The bulk of the decline from the August top occurred through November, before the shutdown and cannot be ascribed to it. The telltale will be the results for January which are due February 12^th.

NFIB Business Optimism

Reuters

Consumer Sentiment Indexes

The University of Michigan Consumer Sentiment Index plunged in January to 91.2 from 98.3 in December. It was the largest one month drop since mid-2011 and brings the index back to the level of September 2016 just prior to the US presidential election and the following two year surge in consumer sentiment.   Here too the decline was not wholly unanticipated but its magnitude may well be related to the government closing which went on for the survey period. The index had reached a 15 year apogee in March at 101.4 and a secondary 2018 high at 100.1 in September.  The 12-month moving average that month set a 17 year record.  

Michigan Consumer Sentiment

Reuters

The Conference Board Consumer Confidence Index has the same general profile. It began climbing in November 2016, reached a 19 year high in October at 137.9 and in the subsequent four months dropped 17.7 points to 120.2 in January, its biggest fall since the financial crisis.  As with  Michigan the steepest part of the decline came in December and January and it is probable if not conclusive to blame the government shutdown.

Conference Board Consumer Sentiment

Reuters

The decline in business outlook is organic. Sentiment saw a very strong surge after the 2016 presidential election. The late summer peaks two years later were ripe for a pullback.  The drop may have been exacerbated by the political conflict in Washington but the rebound in January by the ISM manufacturing index, normally the leading indicator, suggests that the pullback was more a correction than a genuine change in attitude.

Consumer sentiment has been more sensitive to political news but the general pattern is the same. A strong run higher in the aftermath of the election with consolidation in 2018 at levels susceptible to fatigue. The chief difference is that the end of year and January drops are sharper and without the recovery of the manufacturing PMI.

Labor Market Statistics

Labor market planning has a much longer lead time than consumer sentiment and that factor alone was probably sufficient to keep the robust job expansion underway. 

Non-farm payrolls averaged 263,000 in December and January, ADP 238,000. Neither shows any sign of adverse reaction from DC politics.  Wage increases over the year were 3.3% in December and 3.2% in January both the best levels is a decade.

Annual Average Hourly Earnings

Reuters

Jobless claims saw a not unexpected spike to 253,000 in the week of January 26^th , far above the 220,250 4-week moving average, as numbers of laid off government workers sought unemployment benefits. The increase in filings continued into the next week at 234,000.  The long term average remains at a 50 year low.

The Will-o'-the-Wisp Government Shutdown

From the information available it does not appear that the 35 day partial government closure had any effect beyond the immediate Washington vicinity and those workers who were forced to forego their pay.

Business sentiment had peaked and dropped from multi-year highs before the shutdown became a political football in Washington. Consumer sentiment responded to the concerns that a shutdown might stretch out for months but with the closure ended and the labor market booming damage from the dispute should fade quickly.  

Consumer spending is the remaining piece and we will know shortly if US households were worried enough to change their holiday and January white sale habits.

Were another shutdown to happen, the intransigence and division of the capital might become a negative economic factor, since in the eyes of most of the country DC’s responsibility is governance not ideology. Happily neither side seems eager to turn down the shutdown road again. The economic impact of the shutdown will vanish with yesterday’s news.