Instead of trying to predict when the economy will deteriorate in the distant future (which countless experts have tried and failed), we simply look for deterioration among the leading indicators. Instead of predicting the next 10 steps, we seek to predict the next 1-2 steps for the economy.
Here’s a brief summary of the leading economic indicators we track
- Labor market
- Corporate profits
- Financial conditions
- High yield spreads
- Inflation-adjusted new orders
- Heavy Truck Sales
- Earnings revisions
- Yield curve
- Inflation-adjusted retail sales
- Average weekly hours
Special note: trade-related indicators
There’s been a lot of hype recently in the media and social media about the trade war and “global slowdown”. First, a note from Bill McBride (one of my favorite economists because of his accuracy):
Several readers have asked me if I’m on “recession watch”.
The answer is no.
First, a slow growth economy is not a recession. Since the Great Recession ended in 2009, we’ve seen several mini-slowdowns and even a few random quarters of negative GDP growth (but employment and other indicators stayed positive).
Second, the tariffs on goods from China should not have a huge negative impact on U.S. GDP, however the announced tariffs on goods from Mexico appear more significant. I’m relying on the analysis of others to estimate the size of the negative impact, but it doesn’t appear large enough to drag the economy into recession. This could have a significant impact on the auto industry.
There’s a lot to takeaway from such a short answer.
- “Growth slowdown” ≠ recession. Since growth either increases or decreases, growth spends approximately half of its time slowing down (otherwise growth would be infinity!). Even a decrease in growth from 4% to 2% counts as a “slowdown” (while this is still clearly growth!).
- Moreover, “global economy” ≠ U.S. recession. How many times over the past 10 years have we heard the phrase “global recession will lead to contagion in the U.S.!”
- When it comes to predicting the U.S. economy, it’s better to focus on U.S. economic data than to “guess” what Korean export data implies for the U.S. economy.
- Trump’s escalation of a trade war is a real issue. However, it’s important to actually understand how the trade war impacts the U.S. economy instead of panicking HOLY SHIT STOCKS ARE CRASHING WITH THE TRADE WAR!!! RECESSION IS INCOMING! Data > emotions
- The trade war with China does not have a big negative impact on the OVERALL U.S. economy. “Overall” means that some parts of the U.S. economy will be hurt more than others. What the bears and media do is blow these bad parts of the economy out of the proportion, without paying any attention to the good parts of the economy. (Bad news sells).
- The trade war with Mexico is more serious, but it is yet to be seen how this will play out.
Bloomberg has done a terrific job at showing how the trade war impacts GDP.
Keep in mind that U.S. GDP growth is currently at 3.1%. So as Bill McBride has stated, this is not a huge negative for U.S. GDP
You can see how the trade war’s impact on the U.S. economy is worse in 2020 than 2019. Perhaps this means that macro will slowdown in 2020. Wait for the latest data – don’t predict.
Most of the negatives in economic data right now (non-leading indicators) are related to trade or manufacturing. Why don’t we use these indicators? Because they aren’t leading indicators, and they have too many false bearish signals. In a real recession, these tend to turn down long AFTER the recession has started.
The biggest problem most traders have when trying to understand macro is that they don’t rank individual macro indicators based on their importance.
There are thousands of macro indicators. At any given point in time, some will be bullish and some will be bearish. That’s why a good macro trader should know which macro indicators are more important (pay more attention to) and which macro indicators are less important (pay less attention to). That’s the only way to wade through a sea of noise (and the constant bearish lean from financial media certainly doesn’t help!).
Personally, I would rank macro like this.
- Labor market
- Housing market
- Other leading indicators
- Trade, manufacturing
Why this order?
- The labor market DIRECTLY impacts each and every person. Most Americans live paycheck to paycheck, so labor market conditions directly translate to consumer spending.
- The housing market is a relatively small sector of the economy: approximately 4-5% of GDP. However, housing is a very useful leading indicator because it is a big ticket item. Big ticket purchases are the first to fall as the economy slowly slips into a recession. If you’re out of a job, it’s much easier to save money by not spending $300k on a new house than to save money by skimping out on Starbucks.
- Why are trade and manufacturing relatively less important to the U.S. economy than the labor market and housing?
The U.S. is not a manufacturing-centric economy. Manufacturing’s importance has been falling since the 1970s. Manufacturing is much more important in countries such as Germany and China. Once again, you can see how economic indicators are not all created equal, and their importance VARIES from country to country.
In terms of trade, the U.S. economy is a relatively isolated economy. This is because the U.S. economy is much more diverse than most economies, and it is relatively self-sufficient. (A lot of other economies don’t have resource independence, such as China and Japan).
Overall, U.S. macro points to continued economic growth. A recession is unlikely to start within the next few months.
Right now, the most likely start date for a recession is in 2020. If macro remains decent when January 2020 comes around, we will push the most likely start date for a recession even further into the future. Be flexible as the data changes.
Why macro matters
The economy drives corporate earnings, which drives the stock market in the long term. As a result, bull markets usually coincide with economic expansions, and big bear markets usually coincide with recessions.
Since the stock market tends to peak before recessions begin, we need to look at leading economic indicators which also deteriorate before recessions begin.
This doesn’t mean that the stock market’s exact top cannot occur before macro starts to peak. This happened in January 1973, when the stock market peaked a few months before macro started to deteriorate. However, the biggest part of bear markets always occur after macro has deteriorated significantly.
Labor market indicators do not show any significant deterioration right now.
Initial Jobless Claims measures the number of jobless claims filed by people who are seeking to receive jobless benefits. In other words, this measures the number of people who are recently unemployed.
The latest reading for Initial Claims rose from 211k to 215k. More importantly, Initial Claims has been mostly trending sideways over the past 9 months.
In the past, Initial Claims trended higher before a recession began. This is still a positive point for macro, but watch out over the next few months in case Initial Claims trends upwards because Initial Claims is still very low right now.
Continued Jobless Claims measures the number of people who are still filing jobless claims (past the initial claim). In other words, this measures the number of people who remain recently unemployed.
The latest reading for Continued Claims decreased from 1.683 million to 1.657 million. More importantly, Continued Claims has been mostly trending sideways over the past 9 months.
In the past, Continued Claims trended higher before a recession began. This is still a positive point for macro, but watch out over the next few months in case Continued Claims trends upwards because Continued Claims is very low right now.
The Unemployment Rate’s latest reading made a new low for this economic expansion (3.6%).
This is a positive point for macro because in the past, Unemployment trended sideways or upwards before a recession began.
*The Unemployment Rate lags Initial Claims and Continued Claims a little.
KC Fed Labor Market Conditions Index, Momentum Indicator
The KC Fed Labor Market Conditions Index’s latest reading fell from 1.08 to 1.01. More importantly, this index is still above zero.
According to FRED: “A positive value indicates that labor market conditions are above their long-run average, while a negative value signifies that labor market conditions are below their long-run average.” 0 is a necessary but not sufficient requirement for recessions.
The KC Fed Labor Market Conditions Index remains above zero, which means that this “necessary but not sufficient” requirement for recessions has not been fulfilled.
Figures related to corporate profits suggest that while the economic expansion is late cycle, a recession is not imminent.
Unit Profit’s latest reading fell from 99.86 to 99.73. More importantly, Unit Profits has been trending sideways for years.
This is a late-cycle sign for the economic expansion. In the past, Unit Profits usually peaked mid-expansion. While this suggests that the economic expansion is late-cycle, this indicator is not a timing tool. Unit Profits can fall for years before a recession begins.
Inflation adjusted corporate profits’ latest reading fell from 8.21 to 7.90. More importantly, inflation-adjusted corporate profits is still trending higher.
In the past, inflation-adjusted corporate profits trended downwards for several quarters before recessions began.
Indicators related to financial conditions remain relatively loose. This is positive for macro.
Chicago Fed Financial Conditions Credit Subindex
The Chicago Fed Financial Conditions Credit Subindex remains very low at -0.74. More importantly, the Credit Subindex is trending sideways right now.
In a credit-driven economy, the Credit Subindex tends to trend upwards (i.e. tighten) before a recession begins.
Banks’ lending standards
The latest reading for Net Percentage of Banks Tightening Standards fell from 2.8% to -4.2%. More importantly, banks’ lending standards has not trended upwards.
In the past, lending standards tightened for several quarters before a recession began. This is a positive for macro right now.
The latest reading for Delinquency Rate on All Loans has remained the same as last quarter (1.53%). More importantly, the Delinquency Rate has been trending downwards.
In the past, the Delinquency Rate trended higher before a recession began. This is a positive for macro.
High yield spreads
High yield spreads are narrowing right now (i.e. trending downwards).
In the past, high yield spreads tend to widen (trend higher) before recessions began. This is because the average bond market participant is usually more aware of risk than the average stock market participant.
Inflation-adjusted new orders
The latest reading for inflation-adjusted new orders increased from 174.09 to 175.79. More importantly, inflation-adjusted new orders is trending up.
In the past, inflation-adjusted new orders trended downwards before recessions began (this indicator is a component in the Conference Board’s LEI). This is a positive for macro right now.
Heavy Truck Sales
The latest figure for Heavy Truck Sales rose from 0.516 million to 0.547 million. More importantly, Heavy Truck Sales is trending sideways.
In the past, Heavy Truck Sales trended downwards before recessions began. This is a slight positive for macro right now, but watch out in case Heavy Truck Sales starts to fall over the next few months.
Earnings revisions (not the same thing as actual earnings)
U.S. Net Earnings Revisions is now slightly positive
Net Earnings Revisions demonstrates how many companies are increasing their forward earnings estimates minus how many companies are decreasing their forward earnings estimates. In the past, Net Earnings Revisions was negative when recessions started.
This is a necessary but not sufficient condition for recessions, because clearly there are plenty of false signals (see 2011-2016).
The housing market – a key leading segment of the U.S. economy, shows some signs of mild deterioration.
The latest reading for Housing Starts increased from 1168k to 1235k. More importantly, Housing Starts is trending sideways/downwards.
In the past, Housing Starts trended downwards before recessions began. This is a negative for macro.
The latest reading for Building Permits decreased from 1,317k to 1,291k. More importantly, Building Permits is trending downwards.
In the past, Building Permits trended downwards before recessions began. This is a negative for macro.
New Home Sales
The latest reading for New Home Sales decreased from 723k to 673k. More importantly, New Home Sales is trending sideways.
In the past, New Home Sales trended downwards before recessions began. This is neutral for macro.
NAHB Housing Market Index
The latest reading for the NAHB Housing Market Index increased from 63 to 66. After trending downwards in 2018, this leading indicator is starting to recover.
In the past, the NAHB Housing Market Index trended downwards before recessions began. This is a slight negative for macro right now, but could turn positive if it continues to improve for another few months.
Monthly Supply of Houses
The latest reading for Monthly Supply of Houses rose from 5.6 to 5.9. More importantly, this indicator is trending sideways.
The Monthly Supply of Houses looks at the ratio of houses for sale vs. houses sold. In other words, it’s a measure of inventory in the housing market and how hard it is for sellers to sell. The higher this ratio, the tighter the housing market becomes.
In the past, the Monthly Supply of Houses usually (but not always) trended upwards before recessions began. This is neutral for macro right now.
Various sections of the yield curve have either inverted or are close to inverting.
Here is the popular 10 year – 3 month yield curve, which has already inverted.
The 10 year – 3 month yield curve tends to invert 1-1.5 years before a recession begins. The more accurate recession signal occurs when the 10 year – 3 month yield curve steepens after inverting (Fed cuts short term rates as the economy falls into a recession). This has not happened yet.
Here is the 10 year -2 year yield curve, which currently stands at 0.19% (close to inverting).
Like the 10 year – 3 month yield curve, the 10 year – 2 year yield curve tends to invert 1-1.5 years before a recession begins. This has not happened yet.
Overall, the inverting yield curve is a slight negative for macro right now. While this is extremely hyped up in the media and financial media, the yield curve is merely one of many macro points to consider.
Inflation-adjusted retail sales
The latest reading for inflation-adjusted Retail Sales decreased from 2023 to 2013. More importantly, inflation-adjusted Retail Sales has started to trend sideways.
In the past, inflation-adjusted Retail Sales trended sideways before recessions began. This is a slight negative for macro.
Average Weekly Hours of Production and Nonsupervisory Employees: Manufacturing
The year-over-year % change in average weekly hours fell from -1.18% to -1.65%. More importantly, this figure is negative right now.
In the past, this figure was usually in negative territory when recessions began. This is a slight negative factor for macro right now.
This is a necessary but not sufficient condition for recessions, because clearly there are plenty of false signals.
And lastly, a quick word on valuations. The U.S. stock market’s valuations are extremely high right now, as they’ve been for most of this decade. You know it, I know it, everyone knows it.
Here’s the popular Tobin’s Q ratio
Here’s the popular Shiller P/E ratio
However, valuations on their own don’t cause bear markets. Recessions do. Valuations are like the powder, and recessions are like the spark. Higher valuations = more dry powder, which means that when the next recession does hit, the explosion will be bigger.
- High valuations + recessions = big bear market
- Low valuations + recessions = big correction.
So while the bears say “valuations are high, stocks will crash!”, that is merely stating the obvious. Valuations are meant for predicting 10 year forward returns. “Eventually” stocks will always crash. But valuations have very little impact on where the stock market will go over the next month, 6 months, 1 year, 2 years, etc. For example, “valuations are high = poor 10 year forward returns” might mean that the S&P goes up 5 years and then falls 5 years.
These are the S&P 500’s 1 year and 2 year forward returns vs. its valuation (P/E ratio). Notice how the correlation between the stock market’s valuation and 1-2 year forward returns is weak.
R squared = 0.0765 and 0.0646
Shiller P/E consistently peaked at approximately 22 from 1900 – 1994. But from 1994 – present, valuations have been consistently higher. Here’s what happens when you sell the S&P when Shiller P/E reaches to 22. As you can see, a strategy that worked pre-1994 no longer works post-1994.