The Price Relationship That Reliably Signals Stock Market Trends

| About: SPDR S&P (SPY)


The price relationship between non-fuel commodities and gold provides a clear signal of stock market trends and their relative strength.

Empirical evidence of this relationship is convincingly consistent.

This relationship is especially useful as it signals the degree to which confidence in the stock market is supported by underlying economic trends.

Empirical evidence shows that the price relationship between non-fuel commodities and gold on a year-over-year basis signals the durability of stock market trends, positive or negative. The historical accuracy of this relationship is impressive.

Evidence supporting

First some background information. For the real S&P 500 values, I have used the Online Data - Robert Shiller data available here. For the non-fuel commodity prices I am using the Commodity Non-Fuel Price Index published by IndexMundi. I am also using IndexMundi as the source for the price of gold. With one exception identified later, all of the measures - commodity index price, gold price, and real stock market price - are monthly averages.

I have dubbed the model CNF 2 G, which is something of an acronym for commodity non-fuel price to gold price, which is what the relationship is built upon.

To understand the value of the CNF 2 G, look at the three lines within the blue rectangle in Chart 1. The blue trend line that peaks like a mountain is the CNF to G. During this period, when the blue line is above the black line (transition point), the S&P 500 (red line) charged upward. The time period shown here is from November 2012 to December 2014. During that time, the average monthly increase in the S&P 500 was 1.56 percent, which translates to an 18.7 percent annualized increase. (Note: I'm using three-month exponential moving averages for the S&P 500 and CNF 2 G to smooth the trend lines.)

Chart 1

CNF 2 G 1 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

The black line marks the point where the CNF 2 G equals negative 0.050. This value has proven to be a transition point. When the CNF to G stays above negative 0.050, the stock market performs above average, or vice versa.

On the chart below, in the red rectangle, the CNF 2 G (blue line) has dropped below the transition point, the black line. Notice that after this point, the S&P 500 (red line) has been on a roller coaster ride. The monthly average increase in the real S&P 500 during this period, January 2015 to October 2016, has only been 0.1 percent, an annualized rate of 1.2 percent.

Chart 2

CNF 2 G 2 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

In summary, what we are seeing is that when the CNF 2 G has consistently stayed above the transition line, the real S&P 500 has performed above average. When the CNF 2 G is below the transition point, the stock market has performed below average.

This relationship holds up in other periods, but it can be less distinct. In Table 1, periods when the CNF 2 G was equal to or above negative 0.050 are in bold. The far right column shows the average monthly percent increase in the real S&P 500. As can be seen, in the periods where the CNF 2 G was above negative 0.050 the average monthly percentages (BOLD) are consistently higher than the periods when the CNF 2 G was below the transition point.

Table 1 Click to enlarge

The periods above and below the transition line are aggregated in Table 2.

Table 2 Click to enlarge

Even excluding the periods associated with the recessions of 2001 and 2008-2009, during which the stock market had negative returns, there remains a marked difference in the performance of the S&P 500 when the CNF 2 G is above or below the transition point. In other words, even when we look only at periods when the economy is growing and the stock market is increasing, the CNF 2 G discerns between periods that perform above and below average, as shown in Table 3.

Table 3 Click to enlarge

Methodology used

The methodology is surprisingly simple, yet it captures a significant relationship.

In simple terms, the CNF 2 G relationship is calculated by taking the 12-month percent change in the price of gold and subtracting it from the 12-month percent change in price level of non-fuel commodities. This difference signals whether market performance will be above or below average.

Table 4 Click to enlarge

In Table 4, the June 2013 difference of 0.183 would generally indicated that the stock market will perform above average, which it was doing at that time. Whereas, the June 2015 difference, negative 0.064, indicated the stock market would see below average growth, which was also true during that period.

Performance of the model

One aspect of the model is that for periods above or below the transition point, the shortest period of time was nine months (see Table 1). The average number of months during periods when the CNF 2 G was above the transition point was 20 (19.8 to be precise), and for the below periods it was exactly 20 months. This is an important attribute of the CNF 2 G when it is calculated on a year-over-year basis. It captures trends that stretch over multiple months, generally more than a year.

To better understand the model, I'll examine the periods shown in Table 1, starting with the two periods that followed the 2001 recession and the 2008-2009 recession. During these periods, shown in the blue rectangles in Chart 3, the CNF 2 G (BLUE) went above the transition point, indicating that while the economy was improving, there was still some market uncertainty - not unexpected since the economy was coming out of a recession. This uncertainty is demonstrated by the CNF 2 G falling back to the transition point during both periods shown in the boxes. The average monthly increase in the S&P 500 during these periods was 0.19 percent and 0.84 percent, respectively.

During these two periods, the CNF 2 G was signaling that average stock market returns were likely to increase, but it would be a bumpy upward ride, as shown by the S&P 500 performance (red).

Chart 3

CNF 2 G 3 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

Then, interestingly, the two periods shown in Chart 3 were followed by a loss of confidence, shown by the CNF 2 G dropping below the transition point. During these periods, the market did not perform as well (red boxes in Chart 4). The average monthly increases in the S&P 500 were 0.03 percent and 0.76 percent, respectively, less than the monthly averages seen in the prior periods shown in Chart 3.

During these periods, the CNF 2 G is capturing underlying economic and market uncertainty. Accordingly, although the stock market had an overall increase during these periods, it had some steep declines.

Chart 4

CNF 2 G 4 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

The highlighted periods in Chart 4 were followed by the CNF 2 G increasing steadily above the transition line. There was a corresponding improvement in the performance of the stock market (blue boxes in Chart 5). During these periods, while the CNF 2 G stayed at or above positive 0.100 for an extended period, the S&P 500 average monthly increases were 0.74 percent and 1.56 percent, respectively. The prior periods from Chart 4, when the CNF 2 G was below the transition point, were 0.03 percent and 0.76 percent, respectively.

Chart 5

CNF 2 G 5 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

Finally, if we look at the latest phase (red box), through October, the CNF 2 G has been below the transition point. During this period, the average monthly return has only been 0.1 percent.

Of course, the market had a nice rally from February 2016 to August 2016, but from August through October, it flattened. This is what the CNF 2 G was indicating - that there wasn't enough economic strength or market confidence to drive a more prolonged rally.

Chart 6

CNF 2 G 6 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

All of which brings us to recent changes in the CNF 2 G (blue box in Chart 7). In September, the CNF 2 G increased to negative 0.044, just above the transition point. It dropped slightly in October to negative 0.055. In November, based on the recent price of gold and an estimate of commodity price changes, the CNF 2 G reads negative 0.003, a definite improvement from October. Over the past three months, there has been growing confidence in the market. At first glance, the increased optimism may be due to the hope that a new president might focus on economic growth. But, according to data just published by the Bureau of Economic Analysis, when adjusted for inventory valuation and capital consumption, real corporate profits increased by 7 percent in the third quarter. There may be more at play here than just the presidential election.

Chart 7

CNF 2 G 7 Click to enlarge

Sources: IndexMundi and Online Data - Robert Shiller

Even so, part of the November rally seems to be riding on the hope that Trump might pull together an effective pro-growth agenda. Obviously, this rally cannot be sustained solely on what might happen. Whether the economy gets on track has yet to be determined. The CNF 2 G isn't far from the transition point, suggesting that for now market optimism is still quite modest. If the model performs as it should, it will signal if market trends are based on substance or wishful thinking.

What the model captures

Traditionally, investors have moved towards gold as a safe haven asset when there is a substantial threat of market volatility. For example, after the collapse of Lehman Brothers in 2008, investors bought a record 111 tons of gold EFTs in five days.

In contrast, as market confidence increases, investors sell gold holdings, putting downward pressure on its price. In addition, market confidence is highest during periods of strong economic growth. During periods of expansion, producers demand more inputs, putting upward pressure on commodity prices. Thus, when there are periods of improving economic growth and increasing market confidence, the aggregate percent change in non-fuel commodity prices tends to be greater than for gold prices. During such times, the CNF 2 G signals that there is underlying strength in the economy and growing market confidence.

Over the course of a year, the CNF 2 G is capturing the trillions of economic and investment decisions that signal the strength of economic growth and relative confidence in the stock market.

Value of the model

Importantly, the CNF 2 G signals levels of market confidence. When the CNF 2 G remains just above negative 0.050, expect the stock market to perform above average, but the ride will be bumpy. However, when the CNF 2 G moves up to and remains near positive 0.100, this signals high levels of market confidence based on favorable economic conditions. During such periods, the stock market should produce high returns, similar to that shown in Chart 5.

For an investor like me, who isn't trying to pick every high and low, the CNF 2 G tells me exactly what I want to know. As I have followed the CNF 2 G over the past year, it has prescribed a cautious investment strategy. That will be the case until the CNF 2 G makes a strong move above the transition point.

If you believe that you can pick the highs and lows of the market, the CNF 2 G signals whether rallies are likely to be either short term or long term. Positive values near the transition point signal shorter rallies. But, when the CNF 2 G holds near 0.100, it signals bullish periods that are likely to go a year or more.

The CNF 2 G is just one investment tool. However, it provides investors with a vital insight - the extent to which market confidence is supported by underlying economic strength. This is beneficial knowledge for both passive and active investors.

I'm planning to provide regular updates to the CNF 2 G. However, I do have a webpage that shows the current reading of the CNF 2 G.

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.