Are we in race to a recession or to a market bottom? The numbers do not yet look like 2008, but the current trajectories provide sufficient evidence to worry.
Every day we accumulate countless measures, surveys and indicators of economic metrics and corresponding investment behavior. Some observers, such as a recent San Francisco Fed paper take a long view, with the notable observation that expansions and contractions do not necessarily have a preset life expectancy. More conventional assessments of the recession catalysts were summarized in a recent Barron's article. And housing, the key to the 2008 collapse, is still strong. The conventional wisdom is that its lots of things, which does not help isolate the likely triggers for whatever happens next. So despite the abundance of information and analytics, we still don't know what will happen tomorrow. Thus, we are making our decisions based on historical patterns, macro-trends and market price behavior, all interacting with assessments, speculations and fears about near-term economic events - from consumer spending to the next Federal Reserve interest rate decision.
Is business planning anticipating a recession?
There are many hypotheticals, so we wanted to look at robust macro-indicators that reflect both current conditions and planning decisions being made right now, and that will have impact on the next three to six months -the PMI indices of both manufacturing and services sectors of the economy. TrendPointers updated its analysis that compares its proprietary Economic Macro Sentiment Index (TP.EMSI) to the ISM Manufacturing and Non-Manufacturing Purchasing Managers Indices, and its reflection in the S&P 500 stock index.
The TP.EMSI is a composite sentiment-based leading indicator of the 24/7 flow of hundreds of defined economic news, events and related variables from web based sources. EMSI is one of three TrendPointers leading measures and is the sentiment-only measure, i.e. without feedback or other external variables. MacroSentiment Indicators are a rolling weekly index based on a daily harvest of relevant economic news that is coded with an algorithm that accounts for both content and context, and is adaptive to the shifts in cotemporaneous conditions and news. The MacroSentiment Indicators statistically lead metrics such as the PMI, Retail Sales, Housing and the broader stock market by 2 weeks to 2 months, with statistical confidence of 98% and above, and have an effective modeling utility from one week to eight weeks.
For the EMSI/PMI analysis, all the series have been transformed using an exponentially weighted moving average (EWMA) technique and normalized to a standardized scale. With the noisy period-to-period disturbances smoothed out, we are able to more clearly examine the levels and trends among the series over time; both historical and emerging.
Definitions of the variables used in the chart are presented below, followed by a chart itself.
- TP.EMSI: TrendPointers proprietary overall economic macro-sentiment index
- ISM.PMI: ISM Manufacturing Purchasing Managers Index
- ISM.NMI: ISM Non-Manufacturing Purchasing Managers Index
- AVG.ISM: Average of ISM Manufacturing and Non-Manufacturing Purchasing Managers Index
- S&P500: The S&P 500 Index
If we are looking at how "Main Street" business planning, with PMI as the proxy, is anticipating recessionary conditions and a slowdown in business activity, we observe the following in the historical chart:
- Several months before the onset of the recession of 2008, the TP.EMSI economic macro-sentiment index exhibited a steep decline. This was shortly followed by a steep decline in both the PMI indices which quickly transferred into a free fall in the S&P 500 index.
- The TP.EMSI bottom in March 2009 and reversing trend upward. One and a half months later, the PMI indices also reversed trend and in June 2009, the economic recovery began and the S&P 500 stock index began its bull market.
- The PMI indices peaked at the end of 2010, then exhibited a moderate retracement and remained fairly flat between September 2007 and September 2014.
- The S&P 500 stock index during that time had temporary pullbacks in Aug-Sep 2011, May 2012, but remained on a fairly steadfast upward trend that continued through the later part of 2015.
- The TP.EMSI index also had an upward trend while the PMI indices faded. This may help to explain the exuberant mindset of the nation and why the S&P 500 index continued to increase.
- Although the TP.EMSI initially remained on an upward trend, it peaked around March 2013 and reversed to a downward trend through present (Jan 2016).
- The market has largely ignored most warning signs over the past two years, and did not begin to acknowledge economic trouble in the U.S. until August 2015.
In light of the information provided in the EMSI/PMI chart, is the U.S. on the verge of a recession? Unlike prior to the recession of 2008, currently the ISM.NMI has been able to at least partially offset the decline in the ISM.PMI index. Possibly, this has been the reason the market continued higher and ignored the potential recessionary pressures in the economy so persistently. But now that both PMI indices are in marked decline, fears are compounding from relentless negative global economic headlines such as falling crude prices, devaluation of the Yuan, liquidity in the bond market, sovereign debt, and the list goes on. One to two more months will tell the tale if a recession is a high probability, or even started, in 2016. We were all affected in the year up to and after the fall of 2008. Retrospective analysis shows the clues were evident in general and specifically in the TrendPointers EMSI index. We were surprised before, and today no one wants to be the last one out if there's another recession.
Does the data support the market bottom analysis?
If we use PMI as a barometer of macro-economic behavior, how do we link it to the financial markets? Is the market in the long awaited correction or at the beginning of preemptive actions before a recession later this year? Right now, based on the current data, a 2008 repeat seems distant although the negative trend should support serious concerns.
Now both PMI measures are declining and in our view, even 1 to 2 more months of this pattern will provoke, at the very least, a self-fulfilling prophecy of some level of recessionary behavior.
To analyze the behavior of the financial markets, we review price movements in the markets from January 2010 to January 2016, and compare to the PMI analysis:
- Diff200day: the number of NYSE companies above their 200 day moving average minus the number of companies below their 200 day moving average (12 week moving average)
- DiffAdvDec: the number of advancing NYSE companies minus the number of declining companies (12 week moving average)
- TP.RCOBM: TrendPointers proprietary recession/recovery index where the magnitude and duration of a negative index level indicates increased recessionary macro-sentiment.
- S&P500: The S&P500 index.
We can interpret the advances/declines in two ways.
- The glass-half-full interpretation is we are at or very near the market bottom if we compare the bottom of the Diff200day around 10/2011 and can accept its relative proximity to this week.
- The glass half-empty interpretation is that we do not yet see the advance to decline average start to increase, and until we do, markets will not yet see a significant rebound.
What's different and what's the same today?
- In 2008 the economy/markets were "surprised" by the collapse of a core of the economy, due to extralegal behavior.
- Today we are, or claim to be, hyper-vigilant on recession/correction clues, to avoid another surprise.
- But while 2008 was a shadow economic event, today the high visibility catalysts are quite different:
- Oil prices are being managed by premeditated decisions.
- The choice of word from the Federal Reserve.
- So the antecedent conditions to a possible recession are quite different from 2008, but we do not yet know if the outcome will be the same or different, more severe or mild.
The last recession in the United States occurred between December 2007 and June 2009. (Wikipedia) The events that sparked the recession can be linked primarily to the subprime mortgage crisis in the United States which preceded the bursting housing bubble. The price of oil and food were quite high during this time, but the sharp fall of real estate prices and related assets contributed to a global financial crisis.
Today we have a new set of conditions that could spark a recession, and unlike in 2007, every single possible trigger is watched intensely and speculated about endlessly. What makes this different is the origin of the recessionary issues. They are stemming primarily from outside the United States. Although the U.S. is not at the epicenter of the crisis, we are intricately connected to a global economy. A slowdown in China manufacturing, the building of crude inventories, the unwillingness of oil producing companies to curtail oil production, the strength of the Dollar, the devaluation of the Yuan, problems in the EU banks, these series of events may leave no one unscathed. However, based on the evidence to-date as summarized in the EMSI/PMI and Advance/Decline analyses (albeit a simplified perspective), the effect on the U.S. economy does not appear to be in as much trouble today as it was in 2008. By the way, we survived!
As is well known in studies of risk assessment, especially from Kahneman and Tversky's Prospect Theory, we deal with possible negative information far more aggressively than we do with positive information. In the view of TrendPointers, if there is another potential recession, less information will be needed to act preemptively so as not to be guilty of missing the signals once again. As stated earlier, we believe the current economic metrics in the context of TrendPointers Macro Sentiment Indicators requires one to three more months of observed persistence or change to know what the next couple of years will look like.
We are all operating in an environment where there is an excess of real and speculative information, and conventional analytics are excellent at tracing the past but are less successful in anticipating new interactions of expectations, behavior and risk management.
As Keynes said long ago, "Successful investing is anticipating the anticipations of others." And right now the markets are playing this type of anticipatory, cat and mouse game with the available information. One of the problems is, that we don't know exactly who or what to anticipate, because today we have immensely more influential variables than Keynes had to deal with.
Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in SPY, QQQ, GLD, XLE, TLT over 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.