← sources

Reading today’s macro environment with 5 measures, and how I’d position my portfolio

source: reddit (reddit:ValueInvesting)published 2026-10-20 02:53ingested 2026-10-21 02:41

extract

Stocks (especially AI infra) have pulled back from their record highs, 10-year Treasury yields are still above 5%, the Fed raised rates in September for the first time since July 2023, and most Fed officials expect another hike by year-end. So is a recession coming? Is the AI boom really a bubble? And how should you be positioning your portfolio? Here’s where I looked for answers: 1. **Growth × Policy**: A map I use that indicates whether the Fed's rate is high enough to slow the economy. It pairs a growth score (the ISM Services PMI and industrial production) with the real fed funds rate, which is the Fed's policy rate minus core PCE inflation. 2. **Gave's Four Quadrants**: Charles Gave's framework (from Gavekal), which uses two price ratios, S&P 500/oil and gold/Treasuries, to place the economy in a boom or a bust phase, and tell whether it’s inflationary or deflationary. 3. **Credit × Duration**: A map I use that splits bond moves into interest rate risk (the 10-year Treasury yield) and default risk (the high-yield spread, the extra yield junk-rated companies pay over Treasuries). 4. **CAPE Ratio**: Robert Shiller's measure of how expensive U.S. stocks are. It divides the S&P 500's price by its average inflation-adjusted earnings over the last 10 years. I also looked at the Fed model and Shiller’s excess CAPE yield, which corrects the Fed model for inflation. 5. **Bubble Triangle**: William Quinn and John Turner’s framework, which says a bubble needs three sides: marketability (how easily an asset trades), money and credit, and speculation. When a bubble bursts, the damage depends on how tied the bubble is to the rest of the economy, most often through banks. Comparing where these measures/frameworks agree and disagree is more informative than evaluating them on their own. So that's what I did below-- # Growth Is Still Positive Both growth readings are positive for now: * **Growth × Policy**: The ISM Services PMI, a survey of the managers who buy supplies for U.S. services companies, came in at 54.9 in September (above 50 means expansion). That keeps the growth score above zero, even with industrial production up just 1.4% y/y in August. * **Gave's Four Quadrants**: The S&P 500 was worth 85 barrels of oil at the end of September, above its 7-year average of 71. Gave expects growth when stocks outrun oil, since companies are getting more value out of the energy they use. So the economy is still on the growth side, by both the economic data and market prices. # Inflation Is Sticky Inflation shows up in both the data and the price of gold: * **Growth × Policy**: Core PCE inflation, the PCE price index minus food and energy, was 3.0% in August, 1.0% above the Fed's 2% target. * **Gave's Four Quadrants**: At the end of September, an ounce of gold bought 47 shares of IEF, a 7-10 year Treasury ETF, vs. a 7-year average of 26. Gave reads gold beating bonds as the currency losing value, and gold has been beating bonds since March 2020. So Gave's framework places the U.S. in an inflationary boom, where the U.S. has been since November 2023, apart from four weeks in March and April 2026. That's the quadrant where he favors gold, commodities, real estate, value stocks, emerging markets and cyclicals with high fixed costs, and avoids long-duration bonds. High yields don't change that call, since Gave's gold/Treasuries ratio divides the gold price by Treasuries' total return, interest included, and bonds are still losing to gold. # Rates Are Rising Higher rates show up in the bond sell-off, the Fed's hike and gold's lead over bonds: * **Credit × Duration**: Over the 13 weeks to October 9, the 10-year Treasury yield rose from 4.56% to 5.22%, and the high-yield spread widened from 2.69% to 3.15%. 8 of those 13 weeks were driven by rates, with yields rising while spreads held steady or narrowed. On October 5, the 10-year closed at 5.31%, its highest close since May 2002. * **Growth × Policy**: The Fed raised its target by 0.25% on September 16, lifting the fed funds rate to 3.88%. With core PCE at 3.0%, the real fed funds rate is \~0.9%. * **Gave's Four Quadrants**: Rising yields cut IEF's price 4.5% while gold rose 1.8%, so bonds lost to gold. Every U.S. recession since 1970 except COVID began while the real rate was above 1%, and the current \~0.9% is just under that line. Granted, a real rate above 1% doesn't guarantee a recession. Since 1980, the real rate has been above 1% a little more than half the time, including every month from 1995 to 1999. # U.S. Stocks Are Expensive Valuations look stretched on their own and against bonds: * **CAPE ratio**: 40.4x in September 2026, more than double its 18x average since 1881. The CAPE has only been higher in 1999-2000 and August 2026 (month prior). * **CAPE's track record**: Since 1881, a CAPE of 30x or above was followed by -1.0%/year after inflation over the next 10 years on average, and investors lost money 3/5 times. Below 10x, the S&P 500 returned 11.3%/year after inflation, and nobody who bought at those levels lost money. * **Fed model**: Stocks' earnings yield vs. the 10-year Treasury yield. On analysts' forecasts for the next 12 months, the two were about even at the end of September (5.26% vs. 5.29%). On the CAPE's 2.47% earnings yield (1 divided by 40.4), the 10-year's 4.99% September average wins by 2.52%, the widest gap since December 2000. * **Excess CAPE yield**: 0.87% in September, its lowest since April 2002. Since 1881, readings below 2% were followed by 4.1%/year after inflation over the next 10 years on average, vs. 10.4%/year at 6% or more. * **Against TIPS**: The 10-year yield on TIPS, the Treasuries whose principal rises with inflation, averaged 2.64% above inflation in September. That's more than the S&P 500's 2.47% CAPE earnings yield for the first time since at least 2003. * **Concentration**: Info tech was 39.7% of the S&P 500 at the end of September, above its \~33% peak in 2000. So an S&P 500 index fund is now a \~40% bet on tech at a CAPE above 40x. Note: A high CAPE doesn't mean stocks are about to crash. The ratio was above 30x in 80 of the 120 months after September 2016, and the S&P 500 still returned 11.6%/year after inflation over those 10 years. # AI Boom Runs on More Debt The AI boom has all three sides of the bubble triangle: * **Marketability**: Index funds held 52% of U.S. long-term fund assets at the end of 2025, up from 19% in 2010, and they buy each stock in proportion to its size, so the biggest AI stocks get the most of every new dollar. * **Money and credit**: Microsoft, Alphabet, Amazon and Meta are on track to spend more than $730B on capex in 2026, up from $410B in 2025. Over the 12 months to June, the debt of those four plus Oracle grew by an amount equal to 32% of their capex, up from 9% in 2024. Margin debt also hit a record $1.5T in June, \~4.6% of GDP vs. \~3.0% at the dot-com peak. * **Speculation**: AI startups raised 82.7% of the money U.S. venture capital firms invested in 2026 through September, a record share. Quinn and Turner found that tech bubbles with little borrowing, like the dot-com bubble, did little harm to the economy. The ones that ran on borrowing, like the 1840s railway mania and the 1920s, ended in financial crises. With a growing share of AI spending funded by bonds and bank loans, the AI boom is moving toward the ones that ran on borrowing. Note: U.S. household debt was \~66% of GDP in mid-2026, its lowest since 1998, so the borrowing this time is in corporate debt and margin accounts, not mortgages and consumer loans. # Mixed Signals on Cash and Bonds Growth × Policy and the other readings disagree on cash, high-yield bonds and long-term Treasuries: * **Cash**: Growth × Policy is in its risk-on phase, where the growth score is above zero and the real rate is below 1%. That phase says cash lags. But in 2022, when Gave's framework placed the U.S. in an inflationary bust, cash was one of the few assets that held up. * **High-yield bonds**: Risk-on also favors high-yield bonds. But the bonds pay just 3.15% more than Treasuries for taking on default risk, below the 3.62% median since 2019. * **Long-term Treasuries**: Gave avoids long-duration bonds in an inflationary boom. However, if growth slows while the real rate stays under 1%, Growth × Policy moves to a soft landing, which favors long-term Treasuries. # What Would Change the Readings Each of these two paths would flip one of the readings: 1. **2022 path**: Oil spikes, and the S&P 500/oil ratio drops below its 7-year average and stays there, which Gave calls an inflationary bust. The last long inflationary bust was 2022, when the S&P 500 lost 18% and the main U.S. bond index lost 13%, while gold was flat and XLE, the S&P 500 energy ETF, returned 64%. 2. **Credit path**: High-yield bonds' extra yield over Treasuries rises while Treasury yields fall, which Credit × Duration calls a flight to quality. If more of the AI debt has landed on banks by then, the damage would look more like 2008 than 2000. Gave's early warning for the first path has already triggered. The S&P 500/gold ratio has been below its 7-year average since the end of January 2025, and he says that break often comes a few months before the S&P 500/oil break. This time, the S&P 500/oil ratio dipped below its 7-year average for four weeks in March and April 2026, but didn't stay there. From the end of January 2025 to the end of September 2026, the S&P 500 rose 27%, but gold rose 49%. So measured in gold, the S&P 500 fell 15% (from 2.15 oz. of gold to 1.83). # How I'd Position Here's what I'd do with these readings: * **Less in U.S. index funds**: If the CAPE's history holds, U.S. stocks will return less over the next 10 years than the S&P 500's 7.1%/year average after inflation since 1871. * **Value stocks, emerging markets and real assets**: Gave favors all three in an inflationary boom. Growth × Policy favors emerging markets and real estate in its current phase, and value stocks, emerging markets and commodities if the real rate rises above 1%. According to Barclays, the UK (18x CAPE) and China (14x) also trade at less than half the U.S.'s 39x. * **TIPS over nominal Treasuries**: In September, 10-year TIPS yields averaged 2.64% above inflation, more than the S&P 500's 2.47% CAPE earnings yield. Nominal 10-year Treasury yields averaged 4.99%, so over the next 10 years, TIPS beat nominal Treasuries if CPI inflation averages more than 2.35%/year. * **Gold over nominal Treasuries**: Gave favors gold as long as gold/IEF stays above its 7-year average, and the ratio was 47 vs. 26 at the end of September. # What I'm Watching These would change the read the most: * **Fed's October 27-28 meeting and core PCE**: Another 0.25% hike with core PCE at 3.0%, or core PCE below 2.9% with the fed funds rate unchanged, pushes the real rate above 1%. Growth × Policy calls that late cycle, which favors commodities, value stocks, emerging markets and banks over long-term Treasuries and growth stocks. * **S&P 500/oil vs. its 7-year average**: 85 vs. 71 at the end of September. If the ratio falls below its 7-year average and stays there, Gave's framework moves the U.S. into an inflationary bust, like 2022. * **High-yield spread**: 3.15% on October 8. If the spread rises while the 10-year yield falls, investors are selling junk bonds to buy Treasuries, which Credit × Duration calls a flight to quality. During the one in early 2020, long-term Treasuries returned 22% while high-yield bonds lost \~12%. * **Bank-held AI debt**: In September, the Financial Times reported that \~$18B of bank loans for Project Jupiter, a data center Oracle is leasing in New Mexico to serve OpenAI, were being quoted at 89-91 cents on the dollar, and the banks were stuck holding more of the debt than they had planned. Hope you found that helpful! I built a [free macro dashboard](https://stablebread.com/macro) to help me track most of these metrics. And I discuss each of the five measures w/more clarity and detail on my [YouTube channel](https://www.youtube.com/@Fajasy-StableBread).

retained text (truncated)

Stocks (especially AI infra) have pulled back from their record highs, 10-year Treasury yields are still above 5%, the Fed raised rates in September for the first time since July 2023, and most Fed officials expect another hike by year-end.

So is a recession coming? Is the AI boom really a bubble? And how should you be positioning your portfolio?

Here’s where I looked for answers:

1. **Growth × Policy**: A map I use that indicates whether the Fed's rate is high enough to slow the economy. It pairs a growth score (the ISM Services PMI and industrial production) with the real fed funds rate, which is the Fed's policy rate minus core PCE inflation.
2. **Gave's Four Quadrants**: Charles Gave's framework (from Gavekal), which uses two price ratios, S&P 500/oil and gold/Treasuries, to place the economy in a boom or a bust phase, and tell whether it’s inflationary or deflationary.
3. **Credit × Duration**: A map I use that splits bond moves into interest rate risk (the 10-year Treasury yield) and default risk (the high-yield spread, the extra yield junk-rated companies pay over Treasuries).
4. **CAPE Ratio**: Robert Shiller's measure of how expensive U.S. stocks are. It divides the S&P 500's price by its average inflation-adjusted earnings over the last 10 years. I also looked at the Fed model and Shiller’s excess CAPE yield, which corrects the Fed model for inflation.
5. **Bubble Triangle**: William Quinn and John Turner’s framework, which says a bubble needs three sides: marketability (how easily an asset trades), money and credit, and speculation. When a bubble bursts, the damage depends on how tied the bubble is to the rest of the economy, most often through banks.

Comparing where these measures/frameworks agree and disagree is more informative than evaluating them on their own. So that's what I did below--

# Growth Is Still Positive

Both growth readings are positive for now:

* **Growth × Policy**: The ISM Services PMI, a survey of the managers who buy supplies for U.S. services companies, came in at 54.9 in September (above 50 means expansion). That keeps the growth score above zero, even with industrial production up just 1.4% y/y in August.
* **Gave's Four Quadrants**: The S&P 500 was worth 85 barrels of oil at the end of September, above its 7-year average of 71. Gave expects growth when stocks outrun oil, since companies are getting more value out of the energy they use.

So the economy is still on the growth side, by both the economic data and market prices.

# Inflation Is Sticky

Inflation shows up in both the data and the price of gold:

* **Growth × Policy**: Core PCE inflation, the PCE price index minus food and energy, was 3.0% in August, 1.0% above the Fed's 2% target.
* **Gave's Four Quadrants**: At the end of September, an ounce of gold bought 47 shares of IEF, a 7-10 year Treasury ETF, vs. a 7-year average of 26. Gave reads gold beating bonds as the currency losing value, and gold has been beating bonds since March 2020.

So Gave's framework places the U.S. in an inflationary boom, where the U.S. has been since November 2023, apart from four weeks in March and April 2026.

That's the quadrant where he favors gold, commodities, real estate, value stocks, emerging markets and cyclicals with high fixed costs, and avoids long-duration bonds.

High yields don't change that call, since Gave's gold/Treasuries ratio divides the gold price by Treasuries' total return, interest included, and bonds are still losing to gold.

# Rates Are Rising

Higher rates show up in the bond sell-off, the Fed's hike and gold's lead over bonds:

* **Credit × Duration**: Over the 13 weeks to October 9, the 10-year Treasury yield rose from 4.56% to 5.22%, and the high-yield spread widened from 2.69% to 3.15%. 8 of those 13 weeks were driven by rates, with yields rising while spreads held steady or narrowed. On October 5, the 10-year closed at 5.31%, its highest close since May 2002.
* **Growth × Policy**: The Fed raised its target by 0.25% on September 16, lifting the fed funds rate to 3.88%. With core PCE at 3.0%, the real fed funds rate is \~0.9%.
* **Gave's Four Quadrants**: Rising yields cut IEF's price 4.5% while gold rose 1.8%, so bonds lost to gold.

Every U.S. recession since 1970 except COVID began while the real rate was above 1%, and the current \~0.9% is just under that line.

Granted, a real rate above 1% doesn't guarantee a recession. Since 1980, the real rate has been above 1% a little more than half the time, including every month from 1995 to 1999.

# U.S. Stocks Are Expensive

Valuations look stretched on their own and against bonds:

* **CAPE ratio**: 40.4x in September 2026, more than double its 18x average since 1881. The CAPE has only been higher in 1999-2000 and August 2026 (month prior).
* **CAPE's track record**: Since 1881, a CAPE of 30x or above was followed by -1.0%/year after inflation over the next 10 years on average, and investors lost money 3/5 times. Below 10x, the S&P 500 returned 11.3%/year after inflation, and nobody who bought at those levels lost money.
* **Fed model**: Stocks' earnings yield vs. the 10-year Treasury yield. On analysts' forecasts for the next 12 months, the two were about even at the end of September (5.26% vs. 5.29%). On the CAPE's 2.47% earnings yield (1 divided by 40.4), the 10-year's 4.99% September average wins by 2.52%, the widest gap since December 2000.
* **Excess CAPE yield**: 0.87% in September, its lowest since April 2002. Since 1881, readings below 2% were followed by 4.1%/year after inflation over the next 10 years on average, vs. 10.4%/year at 6% or more.
* **Against TIPS**: The 10-year yield on TIPS, the Treasuries whose principal rises with inflation, averaged 2.64% above inflation in September. That's more than the S&P 500's 2.47% CAPE earnings yield for the first time since at least 2003.
* **Concentration**: Info tech was 39.7% of the S&P 500 at the end of September, above its \~33% peak in 2000.

So an S&P 500 index fund is now a \~40% bet on tech at a CAPE above 40x.

Note: A high CAPE doesn't mean stocks are about to crash. The ratio was above 30x in 80 of the 120 months after September 2016, and the S&P 500 still returned 11.6%/year after inflation over those 10 years.

# AI Boom Runs on More Debt

The AI boom has all three sides of the bubble triangle:

* **Marketability**: Index funds held 52% of U.S. long-term fund assets at the end of 2025, up from 19% in 2010, and they buy each stock in proportion to its size, so the biggest AI stocks get the most of every new dollar.
* **Money and credit**: Microsoft, Alphabet, Amazon and Meta are on track to spend more than $730B on capex in 2026, up from $410B in 2025. Over the 12 months to June, the debt of those four plus Oracle grew by an amount equal to 32% of their capex, up from 9% in 2024. Margin debt also hit a record $1.5T in June, \~4.6% of GDP vs. \~3.0% at the dot-com peak.
* **Speculation**: AI startups raised 82.7% of the money U.S. venture capital firms invested in 2026 through September, a record share.

Quinn and Turner found that tech bubbles with little borrowing, like the dot-com bubble, did little harm to the economy. The ones that ran on borrowing, like the 1840s railway mania and the 1920s, ended in financial crises.

With a growing share of AI spending funded by bonds and bank loans, the AI boom is moving toward the ones that ran on borrowing.

Note: U.S. household debt was \~66% of GDP in mid-2026, its lowest since 1998, so the borrowing this time is in corporate debt and margin accounts, not mortgages and consumer loans.

# Mixed Signals on Cash and Bonds

Growth × Policy and the other readings disagree on cash, high-yield bonds and long-term Treasuries:

* **Cash**: Growth × Policy is in its risk-on phase, where the growth score is above zero and the real rate is below 1%. That phase says cash lags. But in 2022, when Gave's framework placed the U.S. in an inflationary bust, cash was one of the few assets that held up.
* **High-yield bonds**: Risk-on also favors high-yield bonds. But the bonds pay just 3.15% more than Treasuries for taking on default risk, below the 3.62% median since 2019.
* **Long-term Treasuries**: Gave avoids long-duration bonds in an inflationary boom. However, if growth slows while the real rate stays under 1%, Growth × Policy moves to a soft landing, which favors long-term Treasuries.

# What Would Change the Readings

Each of these two paths would flip one of the readings:

1. **2022 path**: Oil spikes, and the S&P 500/oil ratio drops below its 7-year average and stays there, which Gave calls an inflationary bust. The last long inflationary bust was 2022, when the S&P 500 lost 18% and the main U.S. bond index lost 13%, while gold was flat and XLE, the S&P 500 energy ETF, returned 64%.
2. **Credit path**: High-yield bonds' extra yield over Treasuries rises while Treasury yields fall, which Credit × Duration calls a flight to quality. If more of the AI debt has landed on banks by then, the damage would look more like 2008 than 2000.

Gave's early warning for the first path has already triggered. The S&P 500/gold ratio has been below its 7-year average since the end of January 2025, and he says that break often comes a few months before the S&P 500/oil break. This time, the S&P 500/oil ratio dipped below its 7-year average for four weeks in March and April 2026, but didn't stay there.

From the end of January 2025 to the end of September 2026, the S&P 500 rose 27%, but gold rose 49%. So measured in gold, the S&P 500 fell 15% (from 2.15 oz. of gold to 1.83).

# How I'd Position

Here's what I'd do with these readings:

* **Less in U.S. index funds**: If the CAPE's history holds, U.S. stocks will return less over the next 10 years than the S&P 500's 7.1%/year average after inflation since 1871.
* **Value stocks, emerging markets and real assets**: Gave favors all three in an inflationary boom. Growth × Policy favors emerging markets and real estate in its current phase, and value stocks, emerging markets and commodities if the real rate rises above 1%. According to Barclays, the UK (18x CAPE) and China (14x) also trade at less than half the U.S.'s 39x.
* **TIPS over nominal Treasuries**: In September, 10-year TIPS yields averaged 2.64% above inflation, more than the S&P 500's 2.47% CAPE earnings yield. Nominal 10-year Treasury yields averaged 4.99%, so over the next 10 years, TIPS beat nominal Treasuries if CPI inflation averages more than 2.35%/year.
* **Gold over nominal Treasuries**: Gave favors gold as long as gold/IEF stays above its 7-year average, and the ratio was 47 vs. 26 at the end of September.

# What I'm Watching

These would change the read the most:

* **Fed's October 27-28 meeting and core PCE**: Another 0.25% hike with core PCE at 3.0%, or core PCE below 2.9% with the fed funds rate unchanged, pushes the real rate above 1%. Growth × Policy calls that late cycle, which favors commodities, value stocks, emerging markets and banks over long-term Treasuries and growth stocks.
* **S&P 500/oil vs. its 7-year average**: 85 vs. 71 at the end of September. If the ratio falls below its 7-year average and stays there, Gave's framework moves the U.S. into an inflationary bust, like 2022.
* **High-yield spread**: 3.15% on October 8. If the spread rises while the 10-year yield falls, investors are selling junk bonds to buy Treasuries, which Credit × Duration calls a flight to quality. During the one in early 2020, long-term Treasuries returned 22% while high-yield bonds lost \~12%.
* **Bank-held AI debt**: In September, the Financial Times reported that \~$18B of bank loans for Project Jupiter, a data center Oracle is leasing in New Mexico to serve OpenAI, were being quoted at 89-91 cents on the dollar, and the banks were stuck holding more of the debt than they had planned.

Hope you found that helpful! I built a [free macro dashboard](https://stablebread.com/macro) to help me track most of these metrics. And 

document fetched 2026-10-21 00:44 · canonical https://reddit.com/r/ValueInvesting/comments/1xagubz/reading_todays_macro_environment_with_5_measures