I reduced my teaching schedule this year to a single MBA derivatives class, which begins in October. This will provide a better balance between teaching and trading and will allow much more time for new research going forward.

In the past two months, I have developed, coded, tested, and implemented two new long-term proprietary strategies: one for commodity futures and one for currency futures. Both look quite promising. The additional time this month also allowed me to smooth the look-back periods for the explanatory variables in the recession model, *which are reflected in this report*. If I have time before I return to teach in October, I hope to test several new variables that I have been considering for the model. Worst case, I will evaluate the new variables after the derivatives class ends.

The following article updates the diffusion index, recession slack index, aggregate recession model, and aggregate peak-trough model through July 2019. The current 21-variable model has a diverse set of explanatory variables and is quite robust. Each of the explanatory variables has predictive power individually; when combined, the group of indicators is able to identify early recession warnings from a wide range of diverse market-based, fundamental, technical, and economic sources.

Several of the explanatory variables are market-based. These variables respond very quickly to changing market conditions and are never revised. This makes the Trader Edge recession model much more responsive than other recession models. The current *and* historical data in this report reflect the current model configuration with all 21 variables.

The Trader Edge diffusion index equals the percentage of independent variables indicating a recession. With the latest changes, there are now a total of 21 explanatory variables, each with a unique look-back period and recession threshold. The resulting diffusion index and changes in the diffusion index are used to calculate the probit, logit, and neural network model forecasts.

The graph of the diffusion index from 1/1/2006 to 8/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly.

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February through April. Upon detailed examination of the individual economic data series, it is clear that the Government shutdown temporarily affected the economic data. The most recent economic data is no longer affected, *but the shutdown was still affecting the look-back data and the resulting trends, which is why I smoothed the data*. Smoothing the look-back data will mitigate the impact of all such data outliers going forward. The number of explanatory variables indicating a recession remained constant at one (4.8%) in July.

Please note that past estimates and index values will change whenever the historical data is revised. All current and past forecasts and index calculations are based on the latest revised data from the current data set.

The Trader Edge recession slack index equals the median standardized deviation of the current value of the explanatory variables from their respective recession thresholds. The resulting value signifies the amount of slack or cushion relative to the recession threshold, expressed in terms of the number of standard deviations. Higher slack values signify larger cushions above recessionary threshold levels. While the *median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

The gray shaded regions in Figure 2 below represent U.S. recessions as defined (after the fact) by the NBER. The *median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

The dark-red, horizontal line at 0.50 standard deviations denotes a possible warning threshold for the recession slack index. Many of the past recessions began when the recession slack index crossed below 0.50. Similarly, many of the past recessions ended when the recession slack index crossed back above 0.0.

In mid-2014, the revised median recession slack index peaked at 1.31, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.54 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.41, but declined in the fall before rebounding at year-end. In early 2019, the median recession slack index dropped below 0.50, but that was due to the temporary and artificial effects of the Government shutdown.

In July 2019, the median recession slack index increased from 0.72 to 0.76. The mean recession slack index increased slightly in June, from 0.57 to 0.69. Note, all of these values reflect the new smoothed look-back data. It is important to recognize that median is more reliable than the mean, because it is not affected by extreme values. The median recession slack index is still slightly above the warning level.

To gain further insight into the slack index, I recently began calculating a derivative value: the percentage of variables with increasing slack each month. The possible values range from zero percent to 100 percent. Due to the monthly volatility, I provide the three-month moving average of the percentage of variables with increasing slack in Figure 3, but I personally monitor the monthly percentages as well.

Slack is a standardized value, so it is directly comparable across all variables. More slack indicates a larger cushion relative to a recessionary environment. As a result, we would like to see as many variables as possible with *increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack remained constant at 49.2% in July. New evidence of economic weakness (or strength) often shows up first in this timely metric.

The ability to track small variations and trend changes over time illustrates the advantage of monitoring the continuous recession slack index. The new slack variable will provide additional insight into the near-term direction of the economy and should be used in conjunction with the median recession slack index.

While it is useful to track the actual recession slack index values and percentage of variables with increasing slack, the diffusion percentages and slack index values are also used to generate the more intuitive probit and logit probability forecasts.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 8/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate decreased from a revised value of 0.2% in June to 0.1% in July. According to the model, the probability that the U.S. is *currently* in a recession is extremely remote.

The peak-trough model forecasts are different from the recession model and are much more responsive. The peak-trough models estimate the probability of the S&P 500 being between the peak and trough associated with an NBER recession. The S&P 500 typically peaks before recessions begin and bottoms out before recessions end. As a result, it is far more difficult for the peak-trough model to fit this data and the model forecasts have larger errors than the recession model.

The Trader Edge aggregate peak-trough model equals the weighted-average of nine different models: the probit and logit models based on the diffusion index, the probit and logit models based on the recession slack index, and five neural network models.

The aggregate peak-trough model estimates from 1/1/2006 to 08/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 8/01/2019 was 5.4%, which declined modestly from last month's revised value of 6.1%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but deteriorated due to the Government shutdown before rebounding in the last few months. The recession risk appeared to increase in January of 2019, but this was largely due to the effects of the Government shutdown.

U.S. recession risk moderated slightly in July. The diffusion index remained constant at one (4.8%), but the mean and median recession slack indices both increased. Both slack indices remain slightly above the early warning threshold. The moving average of explanatory variables with increasing slack remained constant at 49.2% in July. The aggregate recession probability decreased slightly in July (to 0.1%) and the peak-trough recession probability dropped from 6.1% to 5.4%.

Even with the relatively low recession model probabilities, the limited protection offered by the levels of the recession slack indices continues to be a concern, especially with the weak global economy and ongoing trade war.

Hulbert's recent MarketWatch article cited research that used the household equity allocation percentage as a tool for forecasting long-term (10-year) future equity returns. The resulting correlation was so strong (-0.90) that I was compelled to duplicate the research and verify the results myself. I did so and the correlation is correct. It is highly unusual to ever see correlations that high in actual market data. Furthermore, a strong argument can be made for a causal link due to the direct effects of both market valuation and behavioral finance on the household equity allocation percentage.

Based on the most recent data, the regression model indicates that the expected *annual price return* of the S&P 500 index for the next 10 years fell to -1.53%, with an expected drawdown in that period of 39% (from 8/1/2019 levels). Expected price returns are still extremely low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual price return* of the S&P 500 index for the *next 10 years* is materially negative (-3.77%), with an expected drawdown in that 10-year period of 51% (from 8/1/2019 levels).

Overvalued markets can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years.

Unlike human prognosticators, the Trader Edge recession model is completely objective and has no ego. It is not burdened by the emotional need to defend past erroneous forecasts and will always consistently apply the insights gained from new data.

Brian Johnson

Copyright 2019 Trading Insights, LLC. All rights reserved.

]]>Several of the explanatory variables are market-based. These variables respond very quickly to changing market conditions and are never revised. This makes the Trader Edge recession model much more responsive than other recession models. The current *and* historical data in this report reflect the current model configuration with all 21 variables.

The Trader Edge diffusion index equals the percentage of independent variables indicating a recession. With the latest changes, there are now a total of 21 explanatory variables, each with a unique look-back period and recession threshold. The resulting diffusion index and changes in the diffusion index are used to calculate the probit, logit, and neural network model forecasts.

The graph of the diffusion index from 1/1/2006 to 7/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly.

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February through April. Upon detailed examination of the individual economic data series, it is clear that the Government shutdown temporarily affected the economic data. The most recent economic data is no longer affected. The number of explanatory variables indicating a recession declined from three (14.3%) to two (9.5%) in June.

Please note that past estimates and index values will change whenever the historical data is revised. All current and past forecasts and index calculations are based on the latest revised data from the current data set.

The Trader Edge recession slack index equals the median standardized deviation of the current value of the explanatory variables from their respective recession thresholds. The resulting value signifies the amount of slack or cushion relative to the recession threshold, expressed in terms of the number of standard deviations. Higher slack values signify larger cushions above recessionary threshold levels. While the *median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

The gray shaded regions in Figure 2 below represent U.S. recessions as defined (after the fact) by the NBER. The *median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

The dark-red, horizontal line at 0.50 standard deviations denotes a possible warning threshold for the recession slack index. Many of the past recessions began when the recession slack index crossed below 0.50. Similarly, many of the past recessions ended when the recession slack index crossed back above 0.0.

In mid-2014, the revised median recession slack index peaked at 1.31, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.24 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.43, but declined in the fall before rebounding at year-end.

In June 2019, the median recession slack index decreased from 0.58 to 0.51. The mean recession slack index increased slightly in June, from 0.66 to 0.68. It is important to know that median is more reliable than the mean, because it is not affected by extreme values. The median recession slack index is essentially at the warning level. This is particularly troubling given that the effects of the shutdown are no longer artificially depressing the current economic data.

To gain further insight into the slack index, I recently began calculating a derivative value: the percentage of variables with increasing slack each month. The possible values range from zero percent to 100 percent. Due to the monthly volatility, I provide the three-month moving average of the percentage of variables with increasing slack in Figure 3, but I personally monitor the monthly percentages as well.

Slack is a standardized value, so it is directly comparable across all variables. More slack indicates a larger cushion relative to a recessionary environment. As a result, we would like to see as many variables as possible with *increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack declined from 50.8% to 47.6% in June. New evidence of economic weakness (or strength) often shows up first in this timely metric.

The ability to track small variations and trend changes over time illustrates the advantage of monitoring the continuous recession slack index. The new slack variable will provide additional insight into the near-term direction of the economy and should be used in conjunction with the median recession slack index.

While it is useful to track the actual recession slack index values and percentage of variables with increasing slack, the diffusion percentages and slack index values are also used to generate the more intuitive probit and logit probability forecasts.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 7/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased from a revised value of 0.3% in May to 0.4% in June. According to the model, the probability that the U.S. is *currently* in a recession is extremely low.

The peak-trough model forecasts are different from the recession model and are much more responsive. The peak-trough models estimate the probability of the S&P 500 being between the peak and trough associated with an NBER recession. The S&P 500 typically peaks before recessions begin and bottoms out before recessions end. As a result, it is far more difficult for the peak-trough model to fit this data and the model forecasts have larger errors than the recession model.

The Trader Edge aggregate peak-trough model equals the weighted-average of nine different models: the probit and logit models based on the diffusion index, the probit and logit models based on the recession slack index, and five neural network models.

The aggregate peak-trough model estimates from 1/1/2006 to 07/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 7/01/2019 was 9.3%, which declined modestly from last month's revised value of 14.4%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but deteriorated due to the Government shutdown before rebounding in the last few months.

U.S. recession risk moderated in June. The diffusion index dropped from a revised value of three (14.3%) to two (9.5%) in June. The changes in the mean and median recession slack indices were mixed, but both slack indices remain only marginally above the early warning threshold. The moving average of explanatory variables with increasing slack declined from 50.8% to 47.6% in June. The aggregate recession probability increased slightly in June (to 0.4%) and the peak-trough recession probability dropped from 14.4% to 9.3%.

Even with the relatively low recession model probabilities, the limited protection offered by the levels of the recession slack indices continues to be a significant concern.

Hulbert's recent MarketWatch article cited research that used the household equity allocation percentage as a tool for forecasting long-term (10-year) future equity returns. The resulting correlation was so strong (-0.90) that I was compelled to duplicate the research and verify the results myself. I did so and the correlation is correct. It is highly unusual to ever see correlations that high in actual market data. Furthermore, a strong argument can be made for a causal link due to the direct effects of both market valuation and behavioral finance on the household equity allocation percentage.

Based on the most recent data, the regression model indicates that the expected *annual price return* of the S&P 500 index for the next 10 years fell to -1.59%, with an expected drawdown in that period of 39% (from 7/1/2019 levels). Expected price returns are still extremely low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual price return* of the S&P 500 index for the *next 10 years* is materially negative (-3.98%), with an expected drawdown in that 10-year period of 52% (from 7/1/2019 levels).

Overvalued markets can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years.

Unlike human prognosticators, the Trader Edge recession model is completely objective and has no ego. It is not burdened by the emotional need to defend past erroneous forecasts and will always consistently apply the insights gained from new data.

Brian Johnson

Copyright 2019 Trading Insights, LLC. All rights reserved.

]]>

Several of the explanatory variables are market-based. These variables respond very quickly to changing market conditions and are never revised. This makes the Trader Edge recession model much more responsive than other recession models. The current *and* historical data in this report reflect the current model configuration with all 21 variables.

The Trader Edge diffusion index equals the percentage of independent variables indicating a recession. With the latest changes, there are now a total of 21 explanatory variables, each with a unique look-back period and recession threshold. The resulting diffusion index and changes in the diffusion index are used to calculate the probit, logit, and neural network model forecasts.

The graph of the diffusion index from 1/1/2006 to 6/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly.

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February through April. Upon detailed examination of the individual economic data series, it is quite clear that the Government shutdown temporarily affected the economic data. Fortunately, the most recent economic data does not appear to be affected, which means the latest model forecasts are no longer tainted by the questionable data. The number of variables indicating a recession jumped from two (9.5%) to four (19.0%) in May.

Please note that past estimates and index values will change whenever the historical data is revised. All current and past forecasts and index calculations are based on the latest revised data from the current data set.

The Trader Edge recession slack index equals the median standardized deviation of the current value of the explanatory variables from their respective recession thresholds. The resulting value signifies the amount of slack or cushion relative to the recession threshold, expressed in terms of the number of standard deviations. Higher slack values signify larger cushions above recessionary threshold levels. While the *median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

The gray shaded regions in Figure 2 below represent U.S. recessions as defined (after the fact) by the NBER. The *median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

The dark-red, horizontal line at 0.50 standard deviations denotes a possible warning threshold for the recession slack index. Many of the past recessions began when the recession slack index crossed below 0.50. Similarly, many of the past recessions ended when the recession slack index crossed back above 0.0.

In mid-2014, the revised median recession slack index peaked at 1.37, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.27 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.41, but declined in the fall before rebounding at year-end.

In May 2019, the median recession slack index increased from 0.66 to 0.73. The mean recession slack index decreased in April, from 0.71 to 0.68. It is important to know that median is typically more reliable than the mean, because it is not affected by extreme values. The recession slack indices remain only slightly above their early warning thresholds - even after reversing the effects of the shutdown on the economic data.

To gain further insight into the slack index, I recently went back and calculated a derivative value: the percentage of variables with increasing slack each month. The possible values range from zero percent to 100 percent. Due to the monthly volatility, I provide the three-month moving average of the percentage of variables with increasing slack in Figure 3, but I personally monitor the monthly percentages as well.

Slack is a standardized value, so it is directly comparable across all variables. More slack indicates a larger cushion relative to a recessionary environment. As a result, we would like to see as many variables as possible with *increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack declined from 61.9% to 54.0% in May.

New evidence of economic weakness (or strength) often shows up first in this timely metric. The rebound in this variable is somewhat encouraging, but the continued decline in the median and mean slack indices is cause for continued concern, especially with the recent weakness in the equity markets in May.

The ability to track small variations and trend changes over time illustrates the advantage of monitoring the continuous recession slack index. The new slack variable will provide additional insight into the near-term direction of the economy and should be used in conjunction with the median recession slack index.

While it is useful to track the actual recession slack index values and percentage of variables with increasing slack, the diffusion percentages and slack index values are also used to generate the more intuitive probit and logit probability forecasts.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 6/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased from a revised value of 0.1% in April to 0.5% in May. According to the model, the probability that the U.S. is *currently* in a recession is extremely low.

The peak-trough model forecasts are different from the recession model and are much more responsive. The peak-trough models estimate the probability of the S&P 500 being between the peak and trough associated with an NBER recession. The S&P 500 typically peaks before recessions begin and bottoms out before recessions end. As a result, it is far more difficult for the peak-trough model to fit this data and the model forecasts have larger errors than the recession model.

The Trader Edge aggregate peak-trough model equals the weighted-average of nine different models: the probit and logit models based on the diffusion index, the probit and logit models based on the recession slack index, and five neural network models.

The aggregate peak-trough model estimates from 1/1/2006 to 06/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 6/01/2019 was 17.3%, which spiked significantly from last months revised value of 11.2%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but have deteriorated dramatically due to the Government shutdown before reversing in the last few months.

U.S. recession risk definitely increased in May, although recession risk remains relatively low. The diffusion index doubled from two (9.5%) to four (19.0%) in May. The changes in the mean and median recession slack indices were mixed, but both slack indices remain only marginally above the early warning threshold. The moving average of explanatory variables with increasing slack declined from a relatively high level of 61.9% to 54.0% in May. The aggregate recession probability increased slightly in May (to 0.5%) and the peak-trough recession probability jumped from 11.2% to 17.3%.

I explained a few months ago that it was likely that a few of the economic variables were affected by the Government shut-down, which I have now confirmed. While it is encouraging that the model was sensitive to these changes, it also illustrates the importance of qualitatively evaluating exogenous factors that are unknown to the model. The current model readings are representative of the true economic environment. Recession risk increased in May, but the equity market rebound in June (month-to-date) would mitigate some of this risk if the market hold its gains through the end of the month. The limited protection offered by the levels of the recession slack indices continues to be troubling.

Finally, Hulbert's recent MarketWatch article cited research that used the household equity allocation percentage as a tool for forecasting long-term (10-year) future equity returns. The resulting correlation was so strong (-0.90) that I was compelled to duplicate the research and verify the results myself. I did so and the correlation is correct. It is highly unusual to ever see correlations that high in actual market data. Furthermore, a strong argument can be made for a causal link due to the direct effects of both market valuation and behavioral finance on the household equity allocation percentage.

Based on the most recent data, the regression model indicates that the expected *annual price return* of the S&P 500 index for the next 10 years fell to -1.39%, with an expected drawdown in that period of 38% (from 6/1/2019 levels). Expected price returns are still low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual price return* of the S&P 500 index for the *next 10 years* is slightly negative (-3.89%), with an expected drawdown in that 10-year period of 51% (from 6/1/2019 levels).

Overvalued securities can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Unlike human prognosticators, the Trader Edge recession model is completely objective and has no ego. It is not burdened by the emotional need to defend past erroneous forecasts and will always consistently apply the insights gained from new data.

Brian Johnson

Copyright 2019 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 5/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February through April. Upon detailed examination of the individual economic data series, it is quite clear that the Government shutdown temporarily affected the economic data. Fortunately, the most recent economic data does not appear to be affected, which means the latest model forecasts are no longer tainted by the questionable data. The number of variables indicating a recession remained constant at two (9.5%) in April.

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.37, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.27 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.41, but declined in the fall before rebounding at year-end.

In April 2019, the median recession slack index increased from 0.48 to 0.54. The mean recession slack index continued to increase in April, from 0.68 to 0.70. It is important to know that median is typically more reliable than the mean, because it is not affected by extreme values. The good news is that the lows in the past few months were due to the effects of the temporary Government shutdown on several economic series. The bad news is that the recession slack indices remain at or just slightly above their early warning thresholds - even after reversing the effects of the shutdown on the economic data.

To gain further insight into the slack index, I recently went back and calculated a derivative value: the percentage of variables with increasing slack each month. The possible values range from zero percent to 100 percent. Due to the monthly volatility, I provide the three-month moving average of the percentage of variables with increasing slack in Figure 3, but I personally monitor the monthly percentages as well.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack remained constant at 58.7% in April. This rebound is still slightly artificial and is due to reversing the effects of the Government shutdown. It will take another month or so for the 3-month moving average to be representative.

New evidence of economic weakness (or strength) often shows up first in this timely metric. The rebound in this variable is somewhat encouraging, but the continued decline in the median and mean slack indices is cause for continued concern, especially with the recent weakness in the equity markets in May.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 5/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate decreased from a revised value of 1.1% in March to 0.2% in April. According to the model, the probability that the U.S. is *currently* in a recession is extremely low.

The aggregate peak-trough model estimates from 1/1/2006 to 05/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 5/01/2019 was 13.3%, which was unchanged from last months revised value of 13.3%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but have deteriorated dramatically due to the Government shutdown before reversing in the last few months.

U.S. recession risk was relatively stable in April. The diffusion index was unchanged, with two variables (9.5%) indicating a recession in April. The mean and median recession slack indices both increased slightly, but the more reliable median recession slack index remains very close to the early warning threshold. The moving average of explanatory variables with increasing slack remained constant at a relatively high level in April, but the average is still affected by reversing the effects of the government shutdown. The aggregate recession probability decreased slightly in April (to 0.2%), but the peak-trough recession probability remained constant at 13.3%.

I explained two months ago that it was likely that a few of the economic variables were affected by the Government shut-down, which I have now confirmed. While it is encouraging that the model was sensitive to these changes, it also illustrates the importance of qualitatively evaluating exogenous factors that are unknown to the model. The current model readings are representative of the true economic environment. While the April recession model forecasts are more encouraging, recession risk is still slightly elevated and the level of the median recession index is troubling.

Finally, Hulbert's recent MarketWatch article cited research that used the household equity allocation percentage as a tool for forecasting long-term (10-year) future equity returns. The resulting correlation was so strong (-0.90) that I was compelled to duplicate the research and verify the results myself. I did so and the correlation is correct. It is highly unusual to ever see correlations that high in actual market data. Furthermore, a strong argument can be made for a causal link due to the direct effects of both market valuation and behavioral finance on the household equity allocation percentage.

Based on the most recent data, the regression model indicates that the expected *annual price return* of the S&P 500 index for the next 10 years fell to 2.6%, with an expected drawdown in that period of 29% (from 5/1/2019 levels). Expected price returns are still low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual price return* of the S&P 500 index for the *next 10 years* is slightly negative (-0.16%), with an expected drawdown in that 10-year period of 40% (from 5/1/2019 levels).

Overvalued securities can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 4/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February and March. Upon detailed examination of the individual economic data series, it is quite clear that the Government shutdown temporarily affected the economic data. Fortunately, the most recent economic data does not appear to be affected, which means the latest model forecasts are no longer tainted by the questionable data. The number of variables indicating a recession dropped from three (14.3%) to two (9.5%) in March.

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.37, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.27 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.41, but declined in the fall before rebounding at year-end.

In March 2019, the median recession slack index dropped from a 0.60 last month to 0.50. Conversely, the mean recession slack index continued to increase in March, from 0.63 to 0.68. It is important to know that median is typically more reliable than the mean, because it is not affected by extreme values. The good news is that the lows in the past few months were due to the effects of the temporary Government shutdown on several economic series. The bad news is that the recession slack indices remain at or just slightly above their early warning thresholds - even after reversing the effects of the shutdown on the economic data.

To gain further insight into the slack index, I recently went back and calculated a derivative value: the percentage of variables with increasing slack each month. The possible values range from zero percent to 100 percent. Due to the monthly volatility, I provide the three-month moving average of the percentage of variables with increasing slack in Figure 3, but I personally monitor the monthly percentages as well.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack increased from 44.4% last month to 57.1% in March. This rebound is artificial and is due to reversing the effects of the Government shutdown. It will take another couple of months for the 3-month moving average to be representative, although the monthly numbers going forward will be meaningful.

New evidence of economic weakness (or strength) often shows up first in this timely metric. The rebound in this variable is somewhat encouraging, but the continued decline in the median and mean slack indices is cause for continued concern, especially if the rebound proves to be short-lived.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 4/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased slightly from a revised value of 0.6% in February to 0.8% in March. According to the model, the probability that the U.S. is *currently* in a recession is extremely low. However, if the markets do not maintain their recent gains, this probability will increase.

The aggregate peak-trough model estimates from 1/1/2006 to 04/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 4/01/2019 was 12.8%, which decreased by 3.4% from last months revised value of 16.2%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but have deteriorated dramatically due to the Government shutdown before reversing in February.

U.S. recession risk decreased slightly in March. The diffusion index declined from from three variables (14.3%) to two variables (9.5%) in February. The mean recession slack index increased, but the more reliable median recession slack index decreased. Both of the recession slack indices remain at or only marginally above the early warning recession threshold. The moving average of explanatory variables with increasing slack also increased in March, but this was also due to reversing the effects of the government shutdown. The aggregate recession probability increased slightly in March (to 0.8%), but the peak-trough recession probability declined to 12.8%.

I explained two months ago that it was likely that a few of the economic variables were affected by the Government shut-down, which I have now confirmed. While it is encouraging that the model was sensitive to these changes, it also illustrates the importance of qualitatively evaluating exogenous factors that are unknown to the model. The current model readings are representative of the true economic environment. While the March recession model forecasts are more encouraging, recession risk is still elevated and the level of the median recession index is troubling.

Finally, Hulbert's recent MarketWatch article cited research that used the household equity allocation percentage as a tool for forecasting long-term (10-year) future equity returns. The resulting correlation was so strong (-0.90) that I was compelled to duplicate the research and verify the results myself. I did so and the correlation is correct. It is highly unusual to ever see correlations that high in actual market data. Furthermore, a strong argument can be made for a causal link due to the direct effects of both market valuation and behavioral finance on the household equity allocation percentage.

Based on the most recent data, the regression model indicates that the expected *annual price return* of the S&P 500 index for the next 10 years fell to 2.9%, with an expected drawdown in that period of 28% (from 4/1/2019 levels). Expected price returns are still low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual price return* of the S&P 500 index for the *next 10 years* is only slightly positive (+0.07%), with an expected drawdown in that 10-year period of 39% (from 4/1/2019 levels).

Overvalued securities can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 3/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December and January before rebounding in February. Upon detailed examination of the individual economic data series, it is quite clear that the Government shutdown temporarily affected the economic data. Fortunately, the most recent economic data does not appear to be affected, which means the latest model forecasts are no longer tainted by the questionable data. The number of variables indicating a recession dropped from five (23.8%) to three (14.3%) in February, which reversed the temporary effects of the Government shutdown.

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.35, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.27 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.45, but declined in the fall before rebounding at year-end.

In February 2019, the median recession slack index rebounded from a low of 0.49 last month to 0.69. Similarly, the mean recession slack index bounced back from a low of 0.36 last month to 0.64 at the end of February. The good news is that the lows last month were due to the effects of the temporary Government shutdown on several economic series. The bad news is that the recession slack indices are still only marginally above their early warning thresholds - even after reversing the effects of the shutdown on the economic data.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack increased from 38.1% last month to 44.4% in February. This is again due to reversing the effects of the Government shutdown. It will take a few more months for the 3-month moving average to be representative, although the monthly numbers going forward will be meaningful. New evidence of economic weakness (or strength) often shows up first in this timely metric. The rebound in this variable is somewhat encouraging, but the continued decline in the median and mean slack indices is cause for continued concern, especially if the rebound proves to be short-lived.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 3/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate decreased from a revised value of 5.3% in January to 0.4% in February. According to the model, the probability that the U.S. is *currently* in a recession is once again quite low. However, if the markets do not maintain their recent gains, this probability will increase.

The aggregate peak-trough model estimates from 1/1/2006 to 03/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 3/01/2019 was 14.9%, which decreased sharply from last months revised value of 37.7%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but have deteriorated dramatically due to the Government shutdown before reversing in February.

U.S. recession risk decreased significantly in February, primarily as a result of reversing the effects of the Government shutdown. The diffusion index declined from from five variables (23.8%) to three variables (14.3%) in February. Both of the recession slack indices also rebounded sharply, but remain only marginally above the early warning recession threshold. The moving average of explanatory variables with increasing slack also increased in February. The aggregate recession probability dropped significantly in February (to 0.4%), as did the peak-trough probability (to 14.9%).

I explained last month that it was likely that a few of the economic variables were affected by the Government shut-down, which I have now confirmed. While it is encouraging that the model was sensitive to these changes, it also illustrates the importance of qualitatively evaluating exogenous factors that are unknown to the model. The February model readings are representative of the true economic environment. While the February recession model forecasts are more encouraging, recession risk is still elevated.

Based on the most recent data, the regression model indicates that the expected annual price return of the S&P 500 index for the next 10 years rebounded slightly (3.2%), with an expected drawdown in that period of 27% (from 3/1/2019 levels). Expected price returns are still low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual* price return of the S&P 500 index for the *next 10 years* is still negative (-1.0%), with an expected drawdown in that 10-year period of over 50% (-42%) (from 3/1/2019 levels).

*always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 2/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly. However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December. The number of variables indicating a recession dropped from six (28.6%) to five (23.8%) in January. Given the market rebound in January, this is not a surprise. However, while the market-related explanatory variables improved, several economic variables continued to deteriorate. This explains why the diffusion index only declined by one variable (4.8%) in January.

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.35, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.27 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.45, but declined in the fall before rebounding at year-end.

In January 2018, the median recession slack index decreased alarmingly from a revised value of 0.64 to 0.37, falling below the early warning threshold of 0.50 for the first time since early 2016. The mean recession slack index declined from a revised value of 0.40 to 0.35 during the same period. Both the median and mean recession index are now below the early warning threshold.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack increased from 23.8% last month to 39.7% in January. This is consistent with the market rebound, but is inconsistent with the continued decline in the median and mean slack indices. The moving average of variables with *increasing* slack is still very low. New evidence of economic weakness (or strength) often shows up first in this timely metric. The rebound in this variable is somewhat encouraging, but the continued decline in the median and mean slack indices is cause for continued concern, especially if the rebound proves to be short-lived.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 2/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased from a revised value of 8.9% in December to 9.1% in January. According to the model, the probability that the U.S. is *currently* in a recession has increased to almost 10%. However, if the markets do not maintain their recent gains, this probability will continue to increase.

The aggregate peak-trough model estimates from 1/1/2006 to 02/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 2/01/2019 was 40.2%, which increased even further from last months revised value of 34.9%. The current value of 40.2% is the highest peak-trough probability since 4/1/2009.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions improved significantly since early 2016, but have recently deteriorated dramatically. We are now facing another tipping point.

U.S. recession risk continued to increase in January, even with a significant market rebound. The diffusion index declined from from six variables (28.6%) to five variables (23.8%) in January. Despite the decline in the diffusion index, both of the recession slack indices continued their decline and both are now below the early warning threshold for the first time since 2009. The moving average of explanatory variables with increasing slack did increase in January, primarily due to the market-related variables. The aggregate recession probability is now almost 10%, and the peak-trough probability (40.2%) is the highest it has been since the Great Recession.

However, it is possible that a few of the economic variables were affected this month by the Government shut-down. If that was the case, the model readings in the next month or two will be more representative of the true economic environment.

The equity markets rebounded in January, which adds additional price risk to the market on a near term basis. This is particularly troubling when economic conditions continue to worsen. We have reached another inflection point. The next few months will be very telling. Stay tuned...

Based on the most recent data, the regression model indicates that the expected annual price return of the S&P 500 index for the next 10 years has dropped below zero again (-0.11%), with an expected drawdown in that period of 35% (from 2/1/2019 levels). These values worsened again due to the market rebound. Expected price returns are exceptionally low in a historical context, especially given the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual* price return of the S&P 500 index for the *next 10 years* is still negative (-4.3%), with an expected drawdown in that 10-year period of over 50% (-53%) (from 2/1/2019 levels).

*always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 1/1/2019 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly. However, preliminary signs of weakness reemerged in late 2018 and conditions deteriorated rapidly in December. The number of variables indicating a recession jumped from three (14.3%) to five (23.8%) in December; *however, if I update the data through Friday's close (1/18/2019), two of the market-related variables would be back above their recession thresholds due to the market rebound in January. That said, there are still several explanatory variables that are only marginally above their respective recession thresholds.
*

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.35, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.32 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.42, but declined in the fall before rebounding at year-end.

In December 2018, the median recession slack index decreased from a revised value of 0.76 to 0.64. The mean recession slack index declined alarmingly from 0.81 to 0.43 during the same period. The median recession slack indices are still marginally above the early warning threshold of 0.50, but the mean recession slack index dropped below the early warning threshold for the first time since 2009.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack dropped from 31.7% last month to only 25.4% in December. It is extremely unusual for this value to drop this low. This is the lowest 3-month average of variables with *increasing* slack since 2007. New evidence of economic weakness (or strength) often shows up first in this timely metric. The recent trend is very troubling, but could be mitigated if the equity markets continue to rebound in the coming months.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 1/01/2019 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased from 0.5% in November to 3.9%. According to the model, the probability that the U.S. is *currently* in a recession continues to be remote. However, if the markets do not maintain their recent gains, this probability will continue to increase.

The aggregate peak-trough model estimates from 1/1/2006 to 01/01/2019 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 1/01/2019 was 29.6%, which almost doubled last month's revised value of 15.8%. The peak-trough probability would be expected to decrease next month if the equity markets maintain their gains through the end of January.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions have improved significantly since early 2016. The decrease in recession risk has been supported by both market and non-market variables.

U.S. recession risk increased significantly in December. The diffusion index jumped from three (14.3%) to five (23.8%) in December. In addition, both of the recession slack indices declined materially and the mean recession slack index fell below the early warning threshold for the first time since the great recession. The moving average of explanatory variables with increasing slack paints an equally bleak picture. Only 25.4% of the variables had increasing slack, which was the lowest percentage since 2007. The aggregate recession probability is still relatively low, but the peak-trough probability almost doubled in December and is now almost 30%.

The equity markets have rebounded so far in January, which would improve the outlook - at least temporarily. However, we have reached another inflection point. The next few months will be very telling. Stay tuned...

Based on the most recent data, the regression model indicates that the expected annual price return of the S&P 500 index for the next 10 years has improved to +1.05%, with an expected drawdown in that period of 32% (from 1/1/2019 levels). These values have improved recently due to the market correction, but expected price returns of positive 1.05% per year are still quite low, especially with the near-term market, economic, and geopolitical risks.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual* price return of the S&P 500 index for the *next 10 years* is still negative (-2.9%), with an expected drawdown in that 10-year period of almost 50% (-48%) (from 1/1/2019 levels).

*always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 12/1/2018 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly. However, some preliminary signs of weakness have reemerged in late 2018. The number of variables indicating a recession remained at two (9.5%) in November; *however, if I update the data through last night's close (12/20/2018), that number would jump to four, with three more variables within 0.15 standard deviations of their respective recession thresholds. *

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.35, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.32 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.42, but declined in the fall before rebounding at year-end.

In November 2018, the median recession slack index decreased from a revised value of 0.82 to 0.80. The mean recession slack index declined from 0.88 to 0.76 during the same period. I expect these values to drop significantly in December as well. The recession slack indices are still marginally above the early warning threshold of 0.50, but the gap has narrowed significantly and the trend is definitely negative.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack dropped from 38.1% last month to only 31.7% in November. It is unusual for this value to drop this low. New evidence of economic weakness (or strength) often shows up first in this timely metric. The recent trend is very troubling, especially given the continued decline in the equity markets through late December.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 12/01/2018 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate remained constant at 0.2% in November. According to the model, the probability that the U.S. is *currently* in a recession continues to be extremely remote. However, if the markets do not rebound between now and the end of December, this value will jump next month.

The aggregate peak-trough model estimates from 1/1/2006 to 12/01/2018 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 12/01/2018 was 10.4%, which decreased by 1.5% from last month's revised value of 11.9%. As explained above, this is misleading. The market rallied significantly at the end of November, but has declined sharply in December. The peak-trough probability will increase next month (perhaps significantly) if the equity markets do not recover by the end of December.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions have improved significantly since early 2016. The decrease in recession risk has been supported by both market and non-market variables.

U.S. recession risk did not increase in November, but would increase significantly in December if we end the month at current levels. The diffusion index remained constant in November and the recession probabilities remained stable. However, the slack indices declined and the percentage of variables with increasing slack fell precipitously. The peak-trough recession probability remains in low double digits for now, but the recent trend is definitely toward increased weakness.

If the equity markets continue to decline or even remain at current levels, several of the market-based explanatory variables would move into recession territory next month. Stay tuned...

Based on the most recent data, the regression model indicates that the expected annual price return of the S&P 500 index for the next 10 years has dropped below zero (-0.41%), with an expected drawdown in that period of 36% (from 12/1/2018 levels). In other words, the expected price return of the SPX is now negative over the next 10 years, *and* it is likely that an investor would have the opportunity to purchase the SPX at 64% of its current value sometime in the next 10 years. The risk/return trade off for holding periods as short as three years look equally unattractive, albeit with lower correlations.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual* price return of the S&P 500 index for the *next 10 years* is significantly negative (-4.8%), with an expected drawdown in that 10-year period of almost 60% (-54%) (from 12/1/2018 levels).

*always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all domestic and international equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>*and* historical data in this report reflect the current model configuration with all 21 variables.

The graph of the diffusion index from 1/1/2006 to 11/1/2018 is presented in Figure 1 below (in red - left axis). The gray shaded regions in Figure 1 below represent U.S. recessions as defined (after the fact) by the National Bureau of Economic Research (NBER). The value of the S&P 500 index is also included (in blue - right axis).

The U.S. economy flirted with entering a recession in early 2016, which was reflected in the deteriorating economic, fundamental, and especially market-based data. The diffusion index, slack index, and recession probability forecasts all captured the weakening conditions. However, the weakness proved to be temporary and the conditions and recession model forecasts improved rapidly. However, some preliminary signs of weakness have reemerged in late 2018. The number of variables indicating a recession increased from one (4.8%) to two (9.5%) in October.

*median* recession slack index is used in the recession models, I am now including the *mean* recession slack index in the graph as well.

*median* recession slack index is depicted in purple and is plotted against the right axis, which is expressed as the number of standard deviations above the recession threshold. The *mean* recession slack index is depicted in blue and is also plotted against the right axis.

In mid-2014, the revised median recession slack index peaked at 1.35, far above the warning level of 0.50. The recession slack index declined significantly in 2015 and reached a low of 0.32 in February 2016, before rebounding over the next few months. In early 2017, the median recession index peaked at 1.42, but declined in the fall before rebounding at year-end.

In October 2018, the median recession slack index decreased from a revised value of 1.10 to 0.90, which represents a significant 0.20 standard deviation decline in a single month. The recession slack indices are still above the early warning threshold of 0.50, but the gap has narrowed and the trend is negative.

*increasing* slack. Given the diverse nature of the explanatory variables, it is unusual to see more than 60% of the variables with increasing slack or fewer than 40% of the variables with increasing slack. These extreme values are significant and predictive of the near-term direction of economic growth and *often the equity market*.

The 3-month moving average of the percentage of variables with *increasing* slack dropped from 49.2% last month to only 38.1% in October. Only 14.3% of the variables showed *increasing* slack in October. This was the lowest percentage of variables with *increasing* slack in a single month since the Great Recession of 2008-2009. New evidence of economic weakness (or strength) often shows up first in this timely metric. The recent trend is troubling.

The Trader Edge aggregate recession model is the average of four models: the probit and logit models based on the diffusion index and the probit and logit models based on the recession slack index. The aggregate recession model estimates from 1/1/2006 to 11/01/2018 are depicted in Figure 4 below (red line - left vertical axis). The gray shaded regions represent NBER recessions and the blue line reflects the value of the S&P 500 index (right vertical axis). I suggest using a warning threshold of between 20-30% for the aggregate recession model (green horizontal line).

The aggregate recession model probability estimate increased from 0.1% to 0.2% in October. According to the model, the probability that the U.S. is *currently* in a recession continues to be extremely remote.

The aggregate peak-trough model estimates from 1/1/2006 to 11/01/2018 are depicted in Figure 5 below, which uses the same format as Figure 4, except that the shaded regions represent the periods between the peaks and troughs associated with NBER recessions.

The aggregate peak-trough model probability estimate for 11/01/2018 was 10.8%, which increased 4.9% from last month's revised value of 5.9%.

January and February 2016 marked a potential tipping point in U.S. recession risk, but those conditions proved to be temporary. Conditions have improved significantly since early 2016. The decrease in recession risk has been supported by both market and non-market variables.

U.S. recession risk remains relatively low, but increased in October. The diffusion index and recession probabilities all increased, while the slack indices declined sharply. The peak-trough recession probability remains in low double digits for now, but the recent trend is toward weakness.

If the equity markets continue to decline, the effects would be captured by several market-based explanatory variables - long before they are evident in the economic data. Stay tuned...

Based on the most recent data, the regression model indicates that the expected annual price return of the S&P 500 index for the next 10 years has dropped below zero (-1.15%), with an expected drawdown in that period of 38% (from 11/1/2018 levels). In other words, the expected price return of the SPX is now negative over the next 10 years, *and* it is likely that an investor would have the opportunity to purchase the SPX at 62% of its current value sometime in the next 10 years. The risk/return trade off for holding periods as short as three years look equally unattractive, albeit with lower correlations.

I completed a similar historical regression analysis using the "Buffett Indicator", which is the ratio of equity market capitalization to GDP. The correlation is not quite as strong, but is still very significant (-0.74). The Buffett Indicator regression model currently indicates that the expected *annual* price return of the S&P 500 index for the *next 10 years* is significantly negative (-6.4%), with an expected drawdown in that 10-year period of almost 60% (-58.9%) (from 11/1/2018 levels).

Overvalued securities can *always* become more overvalued - especially in the near-term. That said, history offers compelling evidence that bullish equity positions today will face significant headwinds over the coming years. In addition, it is especially troubling that almost all non-US equity markets have suffered dramatic drawdowns in the past 12 months, some as large as 25% to 30%.

Brian Johnson

Copyright 2018 Trading Insights, LLC. All rights reserved.

]]>