US GDP: where is all the growth?

Today’s advance estimate of US GDP shows annualised growth of 0.7% in Q1, above our forecast of 0.2%, but below the consensus estimate. By contrast, business and consumer confidence are at record highs and our US economic sentiment indicator (USESI) points to robust economic activity. The current divergence between hard and soft economic data is unprecedented and is partially explained by transitory factors including seasonal adjustment quirks. But businesses and consumers, while optimistic, want more clarity on the administration’s fiscal and trade policies before loosening the purse strings.

Refresh the chart in your browser | Edit chart in Datastream

It is well known that Donald Trump’s pledge to pursue pro-business policies including cutting taxes, reducing regulations and spending big on infrastructure lifted US equities, the dollar and Treasury yields following his surprise election victory in November last year. But doubts about the President’s ability to make good on these pledges, combined with a run of disappointingly weak economic data releases, which fly in the face of more upbeat surveys, have seen these gains reversed. We still believe that the President will deliver fiscal stimulus later this year, as we explained in a recent note to clients, but what to make of the divergence between the hard and soft data? Is the economy really slowing?

Our US Economic Sentiment Indicator

Consistent with our suite of proprietary economic indicators, we have created a new index to gauge the level of US economic activity implied by business and consumer confidence surveys. The index is called the US Economic Sentiment Indicator (USESI) and uses a technique called principal component analysis. Essentially, we weight together 23 business and consumer confidence surveys to create a single time series – known as the first principal component. We then transform the first principal component so that it has the same mean and variance as actual GDP growth.

Refresh the chart in your browser | Edit chart in Datastream

The USESI should not be viewed as the best predictor of current quarter GDP. That is true not least because the official data will capture large, erratic transactions that simple diffusion indices cannot hope to identify. Rather our USESI should be seen as a measure of underlying economic activity. It tells us where actual GDP growth might be heading if survey measures of business and consumer confidence stay broadly where they are. Our first chart shows that the USESI behaves rather like a moving average of actual GDP growth. It follows a similar cycle, but is much less volatile.

In 2017 Q1, our USESI hit an all-time high of 6.0%. Businesses and consumers are bursting with confidence, and yet the advance estimate of US GDP, published today, shows that the US economy grew by a miserly 0.7% (SAAR) in Q1. Such a large gap between the two measures of economic output is unprecedented.

Beware the official GDP data

One explanation for the divergence is seasonal adjustment issues with the official GDP data. Although the Bureau of Economic Analysis (BEA) tried to address this problem last year, it only tweaked its methodology and revised three years of economic data. A full revamp of its seasonal adjustment methods and a revision to historical data is unlikely to occur before its benchmark revision next year.

Refresh the chart in your browser | Edit chart in Datastream

Accordingly, official economic data still exhibit seasonal bias: since the turn of the century, GDP has grown at an average annualised pace of 1.0% in Q1, compared to average annualised growth of around 2.0% in other quarters.

Unlike the official data, our USESI is free of seasonal bias. It does, however, exhibit significantly higher readings in Q1 than the official output data. The average reading on our USESI in Q1 over the last twenty years is 1.9%, 0.9 percentage points higher than the average for the official GDP growth rates recorded in Q1. Meanwhile, the average readings on the USESI in all other quarters are lower than the average official GDP growth rates in those quarters.

Transitory or troublesome?

Other transitory factors may have depressed the official output figures in 2017 Q1. For example, consumption of electricity and gas services dropped in January and February on the back of unusually warm weather, depressing personal consumption expenditures (PCE). Indeed, the three-month on three-month annualised rate of growth for real PCE was 1.9% in February; excluding consumption of electricity and gas services, this rate was 2.4%. A delay in processing tax rebates for millions of households may have also dampened PCE in Q1, a point noted by some large US retailers.

Nevertheless, some aspects of the hard data cannot simply be dismissed by transitory quirks. For example, according to the BEA, sales of cars and light trucks fell in volume terms by almost 20% on an annualised basis in Q1. Motor vehicle and parts dealers reported a 4% drop in sales, in cash terms, over the same period, according to the Census Bureau. There was also a sizeable fall in the production of automotive products during the first three months of the year.

These drops are at odds with gains in employment, rising wages and improving consumer confidence. However, they are consistent with slowing auto loan growth, rising delinquencies on auto loans and reports from bank loan officers of lower demand and tighter lending standards on auto loans. It would be premature to talk of a credit crunch, although it is possible that the acceleration in consumer credit in recent years has run its course and that slower rates of loan growth are here to stay.

Refresh the chart in your browser | Edit chart in Datastream

Refresh the chart in your browser | Edit chart in Datastream

A little more clarity, please

Stepping back a moment, perhaps the most compelling explanation for the marked divergence between hard and soft indicators at present is that businesses and consumers alike want more detail on the new administration’s economic policies before they loosen the purse strings.

Not only does Paul Ryan’s tax proposal include slashing the corporate tax rate from 35% to 20%, but it also proposes allowing an immediate expensing of investment, removing the tax deductibility of interest expense (thereby making debt financing a lot less attractive) and switching from a ‘global’ to a ‘territorial’ tax system (in which profits earned abroad and brought back to the US would not be taxed). All of these changes would have a massive impact on corporate finance decisions. Businesses are unlikely to commit to significant capital expenditures until the administration provides more clarity on these proposals.

It also remains to be seen where Donald Trump’s US$1 trillion infrastructure spending plans will be targeted, how projects will be auctioned and how they will be financed. Businesses also want clarity on trade policy and whether ‘border adjustment taxes’ will be part of the tax reform. Ultimately, we still think that Donald Trump will succeed in cutting taxes and increasing infrastructure spending later this year, boosting real GDP growth from 1.6% in 2016 to 2.7% this year and 3.5% next year.

_______________________________________________________________________________________

Thomson Reuters Datastream

Financial time series database which allows you to identify and examine trends, generate and test ideas and develop view points on the market.

Thomson Reuters offers the world’s most comprehensive historical database for numerical macroeconomic and cross-asset financial data which started in the 1950s and has grown into an indispensable resource for financial professionals.