Does High CPI Really Weaken the S&P 500?

---

The hypothesis is simple:

When CPI is high, are subsequent S&P 500 returns actually weaker?

This type of question sounds easy to answer, but testing it properly requires several steps.

We need to obtain the macroeconomic data, align it with market data, define the factor, calculate forward returns, and then examine whether the relationship remains meaningful across different parts of the sample.

That is exactly the type of research workflow FactDecode is designed to organize.

For this analysis, CPI is treated as a macro factor and aligned with the S&P 500 market data.

The research setup compares the level of the CPI factor with subsequent S&P 500 performance.

Rather than asking whether a single CPI release caused the market to rise or fall immediately, the focus is on a broader question:

Does the inflation environment at a given point in time contain useful information about future market returns?

This distinction matters.

Short-term market reactions to CPI releases are affected by expectations, positioning, interest-rate pricing, and many other variables. Here, the objective is to examine the historical relationship between the factor itself and forward returns.

The first result is directionally intuitive.

As the CPI factor moves higher, subsequent S&P 500 returns tend to become weaker in parts of the historical sample.

The relationship is visible when the observations are divided into groups based on the CPI factor.

This gives us an initial piece of evidence supporting the common market narrative:

Higher inflation environments have, at times, been associated with weaker subsequent equity-market performance.

But this is where the analysis becomes more interesting.

If CPI alone determined future stock returns, we would expect a clean pattern:

• Lower CPI → stronger forward returns
• Higher CPI → weaker forward returns

The actual data is not that simple.

Some CPI ranges behave differently from neighboring ranges, and the distribution of returns overlaps considerably.

That means we should be careful about turning the result into a rule such as:

“CPI is high, therefore stocks should fall.”

The data does not justify that conclusion.

Instead, CPI appears to be one contributing factor among many.

CPI Matters, but Market Performance Depends on the Regime Around It

A high CPI reading does not exist in isolation.

The market may simultaneously be experiencing:

• Strong or weak economic growth
• Rising or falling interest rates
• Changes in monetary policy expectations
• Different valuation levels
• Different volatility regimes
• Different technical market conditions
• Changes in liquidity and risk appetite

The same CPI level can therefore have a different market meaning depending on the surrounding environment.

For example, inflation accompanied by strong nominal growth may affect equities differently from inflation accompanied by economic contraction.

This is one reason why single-factor analysis is useful as a starting point, but rarely sufficient as a complete market model.

No causal conclusion can be drawn from this analysis alone.

What the data can tell us is whether CPI has historically been associated with differences in forward returns.

That is useful evidence—but association and causation are different questions.

There are several possible mechanisms that could help explain the relationship.

Higher Discount Rates

Persistent inflation can lead markets to price tighter monetary policy and higher interest rates.

Higher discount rates reduce the present value assigned to future cash flows, which can put pressure on equity valuations.

Higher Costs

Inflation can also increase labor, materials, financing, and other operating costs.

Whether companies can pass those costs on to customers depends on the economic environment and pricing power.

Changes in Growth Expectations

Inflation itself may not be the direct driver.

Instead, high inflation may coincide with periods in which monetary tightening, slowing demand, or deteriorating economic expectations become more important to investors.

Multiple Factors Can Be Moving at the Same Time

This is perhaps the most important point.

CPI may matter, but the market is being influenced by many variables simultaneously.

Looking at CPI in isolation tells us only one part of the story.

FactDecode can also inspect how a factor contributes to the output of a machine-learning model using SHAP.

In this analysis, higher CPI observations tend to appear on the side associated with a lower model output, while lower CPI observations are more often associated with a higher output.

This is broadly consistent with the factor analysis above.

But SHAP should not be interpreted as proof that inflation causes lower stock returns.

SHAP explains how the trained model is using a factor within the model.

It helps answer:

How is CPI influencing this model's result?

It does not answer:

Has CPI been proven to cause the market outcome?

That distinction is important when interpreting machine-learning results in financial research.

Another useful question is not simply whether CPI has historically mattered, but where the latest observation sits relative to the distribution used in the analysis.

FactDecode allows the latest factor value to be viewed in the context of the historical research data.

This makes it easier to compare the current environment with the regimes the model has already encountered.

Again, this should not be interpreted as a market forecast.

It is a way to understand where the present observation sits within the evidence used in the research.

This analysis started with a very simple hypothesis:

Does high CPI lead to weaker subsequent S&P 500 returns?

The answer is more nuanced than a simple yes or no.

CPI appears to contain information that the model can use, and higher inflation has been associated with weaker forward returns in parts of the historical data.

At the same time, the relationship is not perfectly linear, and CPI alone cannot explain market behavior.

That naturally leads to the next set of questions.

What happens when CPI is combined with interest rates?

What changes when volatility is added?

Does the relationship depend on the market trend?

Does the result remain similar in validation data?

And which of those factors is actually contributing the most once they are analyzed together?

The CPI example in this article is only one factor.

Real markets involve many variables at the same time: macroeconomic data, price behavior, technical indicators, volatility, interest rates, and other conditions.

FactDecode is a quant research workspace built to help you turn those ideas into testable research evidence.

With FactDecode, you can organize market and macro data, create factors, run AI-assisted analysis, inspect factor contribution and SHAP, and evaluate results using validation data within one research workflow.

Instead of stopping at:

“I think this factor matters.”

you can ask:

“What does the historical evidence actually show?”

If you want to test your own investment hypotheses against market data, you can explore FactDecode here.

[Start Researching with FactDecode]