Origin
The SALSA Index is inspired by the Trump Pressure Index developed by Maximilian Uleer and the macro strategy team at Deutsche Bank. Their research examined how political stress translates into economic pressure and identified conditions under which the administration historically reverses controversial policies.
The term “TACO” (Trump Always Chickens Out) was coined by Robert Armstrong at the Financial Times in May 2025, describing the pattern where aggressive policy announcements are followed by reversals under market and political pressure.
This implementation is an independent, open-source recreation built by Complex AI Institute ( Sam Beni ). It is not affiliated with or endorsed by Deutsche Bank AG or the Financial Times.
The Four Components
The index tracks four macro signals that historically precede policy reversals. Each captures a different dimension of political-economic pressure:
1. Presidential Approval Rating (POLITICAL)
What: Net approval (approve minus disapprove) from RealClearPolitics polling average.
Why: Falling approval is the most direct measure of political cost. When voters punish the president for a policy, reversals follow.
Signal: 4-week change in net approval. Negative = rising pressure.
2. S&P 500 (MARKET)
What: Closing price of the S&P 500 index (^GSPC) via Yahoo Finance.
Why: A falling stock market signals investor panic and creates pressure from the donor class. Administrations historically care deeply about market performance.
Signal: 4-week percentage change. Negative = rising pressure.
3. 10-Year Treasury Yield (BOND MARKET)
What: Daily 10-Year Treasury Constant Maturity Rate (DGS10) from FRED.
Why: Rising long-term yields signal that bond markets are pricing in higher risk. This increases government borrowing costs and squeezes the fiscal outlook.
Signal: 4-week change in percentage points. Positive = rising pressure.
4. 1-Year Inflation Expectations (CONSUMER)
What: University of Michigan 1-Year Inflation Expectations (MICH) from FRED. Released monthly.
Why:Rising inflation expectations mean consumers expect higher prices. This is politically toxic — “kitchen table” economics drives approval more than any other factor.
Signal: Change from prior month. Positive = rising pressure.
Calculation Method
Step 1: 4-Week Changes
For each component, we compute the change over the trailing 4 weeks (20 trading days). For the S&P 500, this is a percentage change. For yields and inflation expectations, this is an absolute change. For approval, this is the change in net approval points.
Step 2: Z-Score Normalization
Each 4-week change is converted to a z-score using the historical mean and standard deviation of that component:
Z-scores express each component in standard deviations from normal. This puts all four components on the same scale, regardless of their original units.
For approval and S&P 500, the z-score is inverted (multiplied by -1) so that deterioration always produces a positive z-score.
Long-term normalization: To prevent the relatively short Trump-era window (14 months) from distorting z-scores, we compute standard deviations from 20 years of market data(2006–2026) and use these as minimum floors. This ensures that a 4-week S&P move of -8% is correctly measured against two decades of market history — including the 2008 crisis, COVID crash, and 2022 rate cycle — rather than just recent months.
All z-scores are capped at ±3.0 to prevent any single component from dominating the composite index.
Step 3: Equal-Weighted Composite
The four z-scores are averaged with equal weight (25% each) to produce a single composite z-score:
Step 4: Scale to 0-100 via Normal CDF
The composite z-score is passed through the standard normal cumulative distribution function (CDF) and multiplied by 100:
This maps the composite smoothly onto a 0–100 scale where 50 represents the historical average. Values above 50 indicate above-average political-economic stress.
🌮 TACO Probability
TACO = Trump Always Chickens Out
Term coined by Robert Armstrong, Financial Times, May 2025
The TACO probability estimates the likelihood of a major policy reversal (tariff pause, trade deal announcement, or significant concession) within the next 14 days. It uses a logistic function:
Where β₀ = -6.0 and β₁ = 0.09. The 50% threshold occurs at index ≈ 67. Reference probabilities:
Note: These coefficients are estimated based on the observed pattern of 10 confirmed TACO events (March 2025 – March 2026), not fitted via formal logistic regression. With limited historical data, this is an approximation rather than a statistically rigorous model.
Scoville Heat Scale
The index value maps to a descriptive heat level inspired by the Scoville scale for chili peppers:
Data Sources
Yahoo Finance via yfinance Python library. Daily closing prices. 20 years of history (2006–present) used for normalization.
FRED (Federal Reserve Economic Data). Daily constant maturity rate. 20 years of history used for normalization.
FRED — University of Michigan Survey. Monthly release, forward-filled to daily. 20 years of history used for normalization.
RealClearPolling Average. Net approval (approve minus disapprove). Daily updates scraped live; historical backfill uses monthly anchors from Silver Bulletin, Gallup, and Pew Research with linear interpolation.
Credits
The Trump Pressure Index was created by Maximilian Uleer and the macro strategy research team at Deutsche Bank. Their work identified the pattern of political-economic stress preceding policy reversals.
The term “TACO” (Trump Always Chickens Out) was coined by Robert Armstrong at the Financial Times in May 2025.
This open-source implementation was built by the Complex AI Institute ( Sam Beni ) to make the concept publicly accessible and transparent.