Seeking Alpha

Forget 3% Growth

by: John Early

Our model and the yield curve suggest the Q2 2.9% year-over-year growth rate may not be exceeded until recovering from the next recession.

Last quarter we thought 2018 would be the best year since 2005. Now 2018 appears on track for 2.7%, which would only be best since 2015.

Our model’s Q3 GDP estimate at 2% is lower than Atlanta Fed at 3.9%, Wall Street Journal Survey at 3.4% and New York Fed at 2.25%.

Next chance of recession still appears to be 2020.

The stock market bubble should continue inflating into 2019.

Expectation the economy is growing at 4% supported by one isolated quarter is bumping up against a long-term downtrend in growth and evidence the next few quarters will grow at less than 3%. While the quarterly growth rate has been trending higher for ten quarters and the annual growth rate trending higher for six, evidence below suggests this surge is not the launch to great prosperity.

The peak growth rate for the current expansion was likely in 2015 and continues the decline in the peak growth rate in place since 1984. If this conclusion is correct, the recent surge in growth is just a fluctuation around a downward trend that will end in recession perhaps in 2020. Discussion below of our quarterly economic model, the yield curve, a demographic model of GDP growth and a labor market indicator support the conclusion the growth rate will decline.

The Model

The Model attempts to find the combination of seven indicators or influences that work together to best predict quarterly GDP. It estimates the relationship of quarterly GDP growth with the quarterly change of these seven factors over rolling-eleven-year periods. The lead time of the factor giving the model the highest correlation with GDP is selected (this may be different than the lead time at which a given factor has the strongest correlation with GDP). As time moves on, factors may come or go from the model. Since July when we last showed the updated model, the personal savings rate for the last 3 years was substantially revised. This revision is the primary reason the Q2 estimate dropped from 5.6% to 4.7% shown above.

The Seven Factors

Personal Savings Rate Inverted (purple line and dots, leading 8 months): A reduction in savings, which means more is spent on consumption, corresponds with a stronger economy about 7 or 8 months later. So this is an inverted or negative relationship. Since the data revision, this series became the least significant in the model and may become a candidate for replacement.

Real retail sales (Orange Line concurrent) grew at a 3.0% annual pace in Q3, a substantial drop from Q2. Retail sales started out so strong and ended so weak in Q3 that our guesstimate real sales will grow at an annual rate of 2.5% each month suggests Q4 would grow at a 1.6% rate from Q3. Retail sales remain the heaviest weighted factor in the model.

Cass Freight index TM From Cass Information Systems, Inc. (Black line and dots, Leading 1 month). Shipments corresponding with Q3 GDP declined 3.3%. This series is not seasonally adjusted. Guesses for the data point used to estimate GDP for Q4 and Q1 2019 are based on the average seasonal change of the previous five years.

Industrial production (green line and dots, leading two months): The months of industrial production indicating Q3 GDP grew at a rate of 2.4%, a big drop from the previous quarter, but still above the average growth for this expansion.

Yield Curve 10-Year – 2-year Treasury Yield (brown line and dots, leading 41 months): The yield curve is often portrayed as the difference between two yields. The yield curve itself has a positive correlation with GDP growth. The best fit of data since 2005 comes with a 26-month lead time.

Historically the optimal lead time has been shorter. By itself the yield curve suggests growth will trend weaker into 2020, but does not indicate a recession.

The brown line in the model above shows the difference in the yield curve from one quarter to the next. In the context of the other variables in the model, the difference in the curve has an inverse relationship with growth with a 41-month lead.

So if the curve flattens or turns negative (inverts) it suggests weaker growth 26 months later. If the rate of decline in the curve accelerates, it will lead the model to project faster growth 41 months later, sort of a rebound from the weaker growth.

Real private inventory (blue line, leading 5 quarters) suggests little change from Q2 to Q3. Inventory indicates faster growth in Q4 and weaker growth in Q1 2018.

Personal Savings Rate (Gold line and dots, leading 31 months): Quarterly growth tends to weaken about 31 months after personal savings weaken. Having a low savings rate hinders growth sustainability. This indicator suggests growth weakens the next two quarters, then strengthens for 2 quarters and then weakens into the first half of 2020.

Q2 2018 Recap: GDP grew at 4.2% - much weaker than the 5.6% we estimated in July. The big savings rate revisions now have the Q2 estimate at 4.7%.

One surprise in Q2 was inflation as measured by the GDP deflator kicked up from 2% to 3.3% and nominal GDP growth was very strong at 7.6%. It was second-best in the expansion.


Our analysis of baby booms and birth-deaths moving through the life cycle suggests growth rising in 2018 and then dropping in 2019 and 2020.

The demographic model above is fitted to the last 25 years. I expect the estimates for 2018–2020 to be useful, but not as accurate as the fitted data would imply. While I currently think the 3.5% estimate for 2018 is too high, I believe the model will be correct that 2018 grows stronger than the 2.2% for 2017 and that the growth rate will decline in 2019 and 2020.

Choosing a 25-year period to analyze is a balancing act. A shorter period gives the best chance picking the lead times with the best correlation to what is going on now. For example, 50 years ago people tended to reach their peak income about age 40. Now peak income tends to be in the early 50s. The downside of a shorter period is the model becomes more subject to over fitting the data.

There are a series of correlations between demographics and GDP growth. About a year after a strong economy births tend to rise, a positive correlation. As parents look after young toddlers who are often sick and need lots of attention, production in the current economy suffers, a negative correlation. As children become more independent and require less attention parents with more time and incentive to provide become more productive, a positive correlation. When young adults reach the late 20s and early 30s they tend to go into debt based on future earnings using mortgages and credit cards. Plus they take risk without the benefits of experience. This tends to weaken growth, a negative correlation. As adults reach their peak income in their early 50s there is a surge in the economy, a positive correlation. Weakness comes as they retire and again start consuming more than they produce, a negative correlation. As they pass on assets and inheritances are freed up to be used in ways that spur growth and family members who were caring for them can become more productive in the economy, a positive correlation.

As baby booms and birth-deaths move through the life cycle the series of influences plays out. Some of the major shifts in the economy are in part demographic echoes of the Civil War and the World War II.

Labor and Growth

Most measures of the labor market lag quarterly GDP growth. One of the few exceptions is the average work week for manufacturing workers. There was a zero percent increase in the average work week from the second quarter to the third quarter which historically is consistent with a 1.5% quarterly growth rate.

Recession Watch

The current expansion at 111 months old is the second-longest and would need to last 29 more months to become the longest. Despite expecting the economy to slow substantially, I see little or no indication of recession in the next year. It is not yet fully indicated a recession appears quite plausible in 2020. On the other hand, if real retail sales and industrial production were to roll over into declines a recession could start much sooner.

Stock Bubble Continues

If the economy continues to expand, the stock market (SPY) should return to making new highs. Typically, the slowing growth final years of an economic expansion coincide with the second-best stock market returns. The best returns, of course, come after a recession and bear market.

The slowing growth final years also have significant corrections. Hoopla about economic strength with expectation of 4% growth has created sentiment almost certain to be disappointed. While my estimate the economy grew at 2% in Q3 could well be low, slowing growth could easily enable sentiment to drive the current stock market correction lower.

Cheerleading about economic strength overstates the significance of the 4.2% second quarter. Prior to 2000, the average growth rate during expansions was over 4%. In this expansion, 4% is more the high end of a fluctuating range. Slower growth could easily weaken sentiment that makes the current stock market weaker.

Even if the stock top were already in, sentiment would probably drive the market up to test the high at some point in coming months. Three months ago when we last updated our economic outlook we called for a new high and got one.

Even as I continue to expect new highs, I also believe a massive bubble in stock prices has been inflated over the last 28 years. Since 1990 the creation of actual wealth, measured by GDP growth, has been about 30% weaker than normal, whereas, the stock market driven by sentiment has risen about 12% faster than normal. This dichotomy cannot be maintained. The next 20 years will likely have below normal stock market returns.

Disclosure: I am/we are long SPY. 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.

Additional disclosure: There is no guarantee analysis of historical data their trends and correlations enable accurate forecasts. The data presented is from sources believed to be reliable, but its accuracy cannot be guaranteed. Past performance does not indicate future results. This is not a recommendation to buy or sell specific securities. This is not an offer to manage money.