Note: If you believe there are economic indicators that should be tracked in the future, you are welcome to reach out to me or leave a comment. The overarching goal is to broadly consider key indicators each month and at the same time selectively narrow in on those indicators that are particularly important and explanatory currently and in the near future.
It is fair to say that economic data over the past year has been disappointing and has weakened from the mediocre pace of growth over the last few years, as the trend of growth has slowed from approximately the 2% range to the 1% range.
Source: Author's calculation from Federal Reserve Bank of St. Louis data.
The YTD average for the four primary coincident indicators has increased at a year over year rate of 1.3% so far this year and each of the four indicators has a growth rate so far in 2016 that is lower than the 2015 growth rate.
Industrial Production peaked in November of 2014 - a full 21 months ago - and is now about 2.1% lower. The underlying causes of the contraction, though, are not all that difficult to determine. The peak in industrial production coincided with the peak in oil prices as well as the beginning of a strong trend upwards in the US dollar.
While falling international demand for exports and declining commodity prices have been poor for industrial production and corporate profits, the other legs of the economy - consumer spending and employment - have held up well and, in the case of consumer spending, has been aided by some of the same forces that have hurt industrial production.
That is makes August weak retail sales number all the more concerning (it declined at an annual rate of 0.5% from July) and has only risen by 1.0% versus last August.
Attempting to divine whether retail sales will continue trending negatively and lead to weaker employment and income growth is not a simple task and it is not possible to know exactly what future months will bring.
When looking at the most recent leading economic indicators it is still too soon to declare that they have rolled over and a recession is on the horizon. Still, most leading economic indicators were weak in August and consumer confidence has now declined for three consecutive months as well as seven months out of the last eight. That does not portend positive things for retail sales.
A weighted average of both of these sets of indicators, leading and coincident, do not point unambiguously in a single direction, but they do appear to have flattened out significantly over the previous year. In August, both sets of indexes declined.
Source: Author's calculations
Various models exist that attempt to determine the likelihood of a recession that is either already happening or likely to occur in the immediate future. Jeremy Piger's model that includes data on the four coincident economic indicators, for example, places an only 0.4% probability on a recession already having begun.
Although looking at a broad swath of indicators does have value, many models have been very successful at predicting recessions by looking at a much more targeted array of metrics. The most used metric in predicting recessions is definitely the term spread, the difference between short term and long term interest rates. The New York Federal Reserve maintains data predicting forward twelve month recession probabilities based on the current spread between 10 year and 3 month interest rates. Right now that probability is less than 10% and the term spread is still more than 100 bps.
Source: New York Federal Reserve
There are good reasons why the term spread is a good indicator of future economic activity. Short term interest rates are heavily dependent on future Federal Reserve activity. If the economy strengthens, short term rates are likely to rise and if the economy weakens they are likely to fall. In turn, we would expect longer term interest rates to track market expectations on average short term rates over the period covered until the bond matures. In other words, an inverted yield curve where short term interest rates are higher than long term interest rates is indicative of market expectations of future monetary loosening.
The problem is that in the low interest rate world of today - where short term rates are up against a lower bound of zero - the term spread is no longer all that predictive, because it is no longer capturing market expectations regarding future economic strength. If the economy were to strengthen then forward interest rate expectations would rise, but if it weakens then short term rates cannot be reduced much further from where they already are. So regardless of whether the economy is likely to strengthen or weaken in the future a probability weighted average of forward interest rate expectations must be higher than the current zero bound where we are currently sitting and the term spread must be positive. If anything, the narrowing of the term spread from 2010 to today (from about 350 bps to 130 bps) is a signal that markets have slowly adjusted to the realization of lower future economic growth.
Recent commentary on the predictive value of the term spread was published by O. Emre Ergungor, an economist at the Cleveland Federal Reserve. Other factors besides the term spread are probably more predictive in the current environment. One such factor is called the "excess bond premium", which unfortunately cannot be directly observed. The excess bond premium starts with credit spreads - the spread between two different credit risks of an identical maturity - and results from dividing that spread into two pieces. The first piece is the difference in default rates between the two issues and the second piece is the excess bond premium.
As an example, if a group of BBB corporate bonds is yielding 200 bps more than a treasury with similar maturity and if the expected default rate on BBB bonds is 100 bps, then the excess bond premium is also 100 bps. This is a form of a risk premium and a rise in the excess bond premium is indicative of future uncertainty or a market prediction that default rates will rise in the future.
Source: Federal Reserve Bank of St. Louis data series BAMLC0A4CBBB and derived from Bank of America Merrill Lynch data.
Early this year, credit spreads widened considerably but have since come back down. According to Ergungor's model (which also includes the term spread, corporate profits, and the return of the S&P 500) recession odds briefly spiked to between 70% and 80%, but are now back to a level of about 30%. In any event, it seems that for the time being watching credit spreads may be the best indicator of the shifting odds of a future recession.
Lost in the mix at times is that compared to the rest of the developed world, the United States economy has performed quite well since 2010. The world at large is definitely in something of an economic funk. The developed world is growing at an anemic rate and the developing world has lost the tailwind of a rising commodity price environment.
The map below shows the growth rates of the fifty largest economies in the world, with those shaded in green growing at greater than 2%, those in yellow at less than 2%, and those in red contracting. The fifty largest economies in the world are responsible for about 94% of global GDP.
Sources: International Monetary Fund, World Bank, and OECD.
On the whole, Q2 global growth was ~2.4%, but it is somewhat worrying that so much of global growth is coming from a few countries. China, the United States, and India accounted for about 2/3rds of Q2 growth, with China alone responsible for more than 40%. For that reason, I plan to focus on the state of the Chinese economy in next month's update along with a refreshed look at the progression of the US economy.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.