Presidential Partisanship and Sectoral ETF Performance in U.S. Equity Markets

Abstract

This study investigates how U.S. presidential political leadership affects financial market performance at the sector level, offering a novel contribution to the literature that has largely focused on aggregate market indices. While prior research documents partisan effects on overall stock returns, little is known about how different sectors respond to changes in political leadership. Using sector-specific exchange-traded funds (ETFs) categorized by the Global Industry Classification Standard (GICS), we examine sectoral return patterns and volatility under Republican and Democratic presidencies. This study contributes to the growing intersection of finance and political economy by providing a nuanced, empirical understanding of sectoral behavior across political cycles. The results offer valuable insights for investors, portfolio managers, and policymakers, enhancing their ability to anticipate sector-level risks and opportunities under changing political leadership.

1. Introduction

The dynamic relationship between political leadership and financial market performance has long captivated the interest of scholars, policymakers, and investors. As political power shifts—particularly through changes in presidential administrations, these transitions shape investor sentiment, recalibrate regulatory expectations, and ultimately influence asset prices. Seminal contributions to this field, such as Santa-Clara and Valkanov (2003) and more recently Pastor and Veronesi (2020), have documented consistent differences in U.S. stock market returns depending on which political party occupies the White House. These findings underscore political cycles as a vital dimension of financial market dynamics, signaling that politics and markets are tightly intertwined.

Despite this growing understanding, much of the existing literature remains anchored in broad market indices like the S&P 500 or Dow Jones Industrial Average. While these aggregates offer valuable snapshots of overall market behavior, they risk overlooking the important heterogeneity embedded within the economy. Different sectors respond uniquely to policy changes and regulatory shifts depending on their exposure and sensitivity to government priorities. For instance, energy companies often benefit from deregulation and fossil fuel subsidies typically favored under Republican administrations, whereas healthcare and renewable energy sectors tend to gain from Democratic leadership that emphasizes public health initiatives and environmental sustainability.

Recognizing this gap, our study aims to delve deeper by examining sector-specific financial performance under Democratic and Republican presidencies by leveraging exchange-traded funds (ETFs) classified by the Global Industry Classification Standard (GICS). These ETFs provide transparent, investable proxies for economic sectors, enabling consistent measurement of returns and volatility over time and across political regimes. This sector-level perspective allows us to uncover whether certain industries systematically outperform or underperform contingent on partisan control of the presidency.

For instance, during Republican administrations, investors may benefit from overweighting energy, defense, and financial sector ETFs, which historically respond positively to deregulation and increased defense budgets. Conversely, during Democratic administrations, technology, renewable energy, and healthcare ETFs tend to show stronger performance, reflecting policy emphasis on innovation, clean energy, and healthcare reform. In transitional or uncertain periods, defensive sectors such as utilities and consumer staples provide relative stability, making them attractive allocations for risk-averse investors.

By shifting the analytical lens from aggregate market indices to sector-specific ETFs, we capture a more granular and realistic picture of how political leadership influences different facets of the economy. This nuance is crucial for investors aiming to manage political risk or optimize sector allocations ahead of electoral outcomes, as well as for policymakers seeking to anticipate the economic repercussions of their platforms.

Moreover, prior studies may be confounded by extraordinary market events—such as the 2008 global financial crisis or the COVID-19 pandemic—that can obscure the direct effects of political leadership on market performance. To address this issue, we employ a two-pronged empirical strategy: a Baseline Analysis incorporating the full dataset including crisis periods, and a Filtered Analysis that excludes episodes of extreme volatility and drawdowns to isolate more structural relationships. This approach strengthens the robustness of our results by distinguishing between market reactions driven by politics and those triggered by broader systemic shocks.

Through this nuanced exploration, our study contributes to a richer understanding of how political leadership interacts with financial markets at the sector level, offering valuable insights for academia, investors, and policymakers alike.

2. Literature Review

This chapter reviews the scholarly work that intersects two crucial themes for our study: the influence of political leadership on financial markets and the evolving role of exchange-traded funds (ETFs) in contemporary investing.

2.1. Political Leadership and Stock Market Performance

The interplay between U.S. presidential cycles and equity market outcomes has been a fertile area of research. Santa-Clara and Valkanov (2003) famously introduced the “presidential puzzle,” demonstrating that stock returns tend to be significantly higher during Democratic administrations. This aggregate-level observation has sparked ongoing debate regarding the underlying drivers of political effects on markets.

Building on partisanship, Montone (2022) shifts the focus to presidential approval ratings, finding that periods of high net disapproval—regardless of political party—predict lower future returns. This suggests that investor sentiment and perceived policy uncertainty, rather than party labels alone, play a central role in shaping market responses. Complementing this perspective, Leblang and Mukherjee (2005) link partisan macroeconomic expectations, such as anticipated inflation under left-leaning governments, to trading behavior. They show that both return levels and volatility vary systematically with the incumbent party, with right-leaning governments typically inducing higher trading volume and volatility. Together, these studies highlight how political orientation and public perception influence market dynamics beyond aggregate returns.

Volatility in response to political uncertainty is a recurring theme. Mnasri and Essaddam (2021) document spikes in implied volatility around U.S. presidential elections, particularly when an opposition candidate’s victory appears likely, supporting the Election Uncertainty and Political Uncertainty hypotheses. Similarly, Ma et al. (2024) show that firm-level political risk exposure in China increases the likelihood of stock price crashes, illustrating that political shocks can propagate through corporate channels.

While much of the literature treats the stock market as a monolithic entity, sector-specific effects remain underexplored. Alim et al. (2024) provide evidence from Pakistan that political stability affects sectoral returns and volatility, with negative shocks disproportionately affecting some industries. These findings suggest that sensitivity to political risk varies across sectors and highlight the importance of analyzing political effects at a more granular level.

Recent work also emphasizes the bidirectional relationship between politics and markets. Bonaparte (2025) uses panel data from 60 countries to show that institutional structures shape market behavior, finding that presidential systems exhibit lower volatility than parliamentary systems due to greater political stability. Extending this perspective, Crane et al. (2024) provide evidence from the U.S. that stock market performance can influence political outcomes. They find that counties with higher levels of market participation are more likely to support the incumbent party when markets perform well, revealing a novel channel through which market fluctuations can shape election outcomes. These studies collectively underscore that financial markets not only respond to political dynamics but can also exert real political influence.

Internationally, partisan effects are context-dependent. Bialkowski et al. (2007) show that the U.S. partisan-return relationship does not generalize to 24 global markets, highlighting the role of institutional and contextual differences. Similarly, Castaño et al. (2024) find no partisan effects in Spain but note that volatility spikes occur during government transitions, emphasizing uncertainty rather than ideology as a driver.

In emerging markets, political cycles influence markets differently. Yiadom et al. (2024) find that election campaigns in African countries tend to boost market performance, whereas post-election regime changes depress returns, suggesting that predictability, rather than ideology, reassures investors. Souffargi and Boubaker (2024) show that democratic transitions in Tunisia improve market returns, while uprisings and partisan conflicts negatively affect performance. Together, these studies highlight the primacy of political stability in shaping investor expectations worldwide.

Overall, the literature illustrates that the relationship between politics and markets is complex, shaped by institutional structures, investor sentiment, and market participation. While previous research has examined aggregate market responses, sector-specific dynamics and the feedback effects from markets to politics remain important areas for further exploration—gaps that this study seeks to address.

2.2. Exchange-Traded Funds: Structure, Performance, and Behavior

ETFs have transformed investing by offering cost-effective, liquid, and transparent exposure to broad and niche asset classes alike. Since the 1993 launch of the first U.S.-listed ETF—the SPDR S&P 500—ETF markets have grown exponentially, now spanning nearly every sector, region, and strategy, with increasing emphasis on thematic, ESG, and AI-driven products.

Lettau and Madhavan (2018) describe ETFs as a financial innovation that combines tax efficiency, low fees, and intraday tradability, noting that despite concerns, systemic risk is largely mitigated by regulatory oversight.

Examining geographic disparities, Tarassov (2016) highlights limited ETF availability in Russia, pointing to high investor costs due to dependence on mutual funds, and calls for better investor education and infrastructure in emerging markets.

In China, Wu et al. (2021) compare ETFs with traditional index funds, showing that ETFs outperform both before and after fees, especially during volatile periods, where ETFs demonstrate more decentralized investor behavior. However, diversification benefits may be overstated in less mature markets.

Valadkhani and Moradi-Motlagh (2023) analyze 110 U.S.-listed ETFs, identifying a subset of “super-efficient” funds in technology and healthcare sectors that consistently outperform peers regardless of risk preferences, challenging assumptions that passive investing yields average returns.

Investor behavior research by Lee et al. (2021) finds that home-country investor attention strongly influences returns of single-country ETFs, with preferences skewed toward countries institutionally similar to the U.S., highlighting the roles of information asymmetry and familiarity bias.

Joshi and Dash (2024) provide a comprehensive bibliometric review of ETF research, noting a surge in publications post-2020 focused on volatility, tracking errors, and ESG integration. They emphasize that ETF research remains heavily concentrated in developed markets, calling for greater attention to emerging economies.

2.3. Bridging the Gap: Political Influence on Sector ETFs

While prior research has explored the influence of political cycles on financial markets and has separately examined the growing role of exchange-traded funds (ETFs), there remains a lack of systematic investigation into how presidential leadership directly shapes sector-specific ETF performance. This omission is notable, given that sector ETFs provide a transparent and accessible window into investor expectations about the future of different parts of the U.S. economy under varying political administrations.

Our study makes a new contribution by explicitly bridging these two strands of literature. Instead of analyzing broad market indices, which often obscure sectoral differences, we use sector ETFs as practical proxies for distinct slices of the U.S. equity market. This approach enables us to capture heterogeneity across sectors—for example, the potential for energy, healthcare, or technology to respond differently to partisan policy agendas. By doing so, we extend the discussion beyond aggregate stock market effects and provide evidence that political cycles may have uneven impacts across the economic landscape.

3. Data and Methodology

Our data is sourced from Yahoo Finance. We collected historical monthly price data for a selection of sector-based exchange-traded funds (ETFs) and computed monthly returns using the adjusted closing prices, which account for dividends and stock splits. Specifically, the monthly return is calculated as R_t = (P_t − P_t−1)/P_t−1, while P_t and P_t−1 denote the adjusted closing prices at the end and beginning of the period, respectively. Please note that Yahoo Finance provides monthly adjusted prices as of the first trading day of each month. Accordingly, the beginning-of-period price corresponds to the first trading day of the current month, while the end-of-period price corresponds to the first trading day of the following month.

Table 1. Detailed Information and Data History of ETFS used in the Study. This table presents detailed information for each ETF, including its relevant historical data.

Ticker	Description	Data History
SPY	SPDR ETF for S&P 500 Index	01/93–03/25
QQQ	Invesco ETF for QQQ100 Index	03/99–03/25
SPDR sector ETFs
XLE—energy	SPDR Energy Sector ETF	12/98–03/25
XLK—Information Technology	SPDR Information Technology Sector ETF	12/98–03/25
XLI—Industrials	SPDR Industrial Sector ETF	12/98–03/25
XLF—Financials	SPDR Financials Sector ETF	12/98–03/25
XLC—Communication Services	SPDR Communication Services Sector ETF	06/18–03/25
XLY—Consumer Discretionary	SPDR Consumer Discretionary Sector ETF	12/98–03/25
XLP—Consumer staples	SPDR Consumer Staples Sector ETF	12/98–03/25
XLB—Materials	SPDR Materials Sector ETF	12/98–03/25
XLV—Healthcare	SPDR Healthcare Sector ETF	12/98–03/25
XLU—Utilities	SPDR Utilities Sector ETF	12/98–03/25
XLRE—Real Estate	SPDR—Real Estate Sector ETF	10/15–03/25

provides detailed information on the ETFs included in this study, including their sector classification, ticker symbols, and the time span of the available data used in the study.

To examine whether sectoral returns differ significantly under Republican versus Democratic presidential leadership, we first conduct two-sample t-tests comparing the average returns of each ETF across the two political regimes. This allows us to assess whether the mean return for a given sector-based ETF is statistically different depending on the party in the office.

In addition, we implement regression analysis to further investigate the impact of presidential leadership on ETF returns. A dummy variable, Rdummy, is constructed to indicate periods of Republican presidential leadership (Rdummy = 1 for Republican administrations; 0 otherwise). Each ETF return is then regressed on Rdummy, the market return proxied by the SPDR S&P 500 ETF (SPY), and an interaction term (SPY × Rdummy). This interaction term captures whether the influence of market-wide movements on sector-specific returns varies by political leadership.

We estimate two regression models—referred to as Model 1 and Model 2. These models are designed to evaluate both the direct effect of presidential leadership on the ETF returns sector and the indirect effect via changes in the relationship between sector returns and overall market returns.

$$R_i= \alpha +{\beta }_M\times R_M+{\beta }_{Rdummy}\times Rdummy$$

(1)

$$R_i= \alpha +{\beta }_M\times R_M+{\beta }_{Rdummy}\times Rdummy+{\beta }_{interactive}\times (R_M\times Rdummy)$$

(2)

The use of both t-tests and regression analysis provides complementary insights: t-tests offer a straightforward comparison of mean differences across political regimes, while regression models allow us to control market-wide movements and isolate the incremental impact of presidential partisanship. By combining these approaches, we ensure that our findings are not only descriptive but also account for broader risk factors that might otherwise confound the results.

To strengthen inference, we also confront the confounding influence of extreme market events—such as the 2008 financial crisis and the COVID-19 pandemic—by adopting a twofold methodology:

• Baseline Analysis: Includes the full dataset, capturing the entire historical context including crises.
• Filtered Analysis: Excludes periods of extreme volatility and drawdowns to isolate structural political effects.

This approach enables a robust examination of political effects on sectoral performance under varying market conditions, disentangling direct political impacts from broader systemic shocks. In doing so, we contribute a more refined understanding of the interplay between politics and financial markets at the sectoral level.

4. Empirical Results with Full Data Sample

In this chapter, we present the empirical findings of our analysis.

Table 2. Descriptive Statistics for Various ETFs. This table presents the descriptive statistics of various ETFs used in the study, including mean, sample standard deviation, and minimum and maximum value as well as the number of observations.

All Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
	01/93–03/25	03/99–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	06/18–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	11/15–03/25
Mean	0.89%	0.97%	0.88%	0.88%	0.82%	0.64%	0.98%	0.87%	0.61%	0.78%	0.73%	0.69%	0.66%
Standev.S	4.35%	6.79%	7.36%	6.62%	5.48%	6.22%	5.79%	5.64%	3.64%	6.11%	4.12%	4.47%	5.16%
Min	−16.03%	−26.20%	−35.80%	−24.90%	−19.25%	−25.07%	−13.91%	−17.36%	−11.98%	−21.95%	−14.51%	−13.89%	−15.73%
Max	13.36%	24.32%	33.68%	24.86%	19.23%	22.89%	15.07%	19.46%	9.72%	25.12%	13.13%	12.37%	12.47%
Obs	387	313	316	316	316	316	82	316	316	316	316	316	114
Republican Presidential Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Mean	0.28%	0.42%	0.28%	0.41%	0.30%	−0.09%	0.96%	0.46%	0.33%	0.63%	0.51%	0.48%	0.72%
Standev.S	4.58%	7.29%	7.89%	7.27%	5.43%	5.79%	6.11%	5.56%	3.42%	5.71%	4.25%	4.35%	4.33%
Min	−16.03%	−26.20%	−35.80%	−24.90%	−19.25%	−21.70%	−12.84%	−17.36%	−11.98%	−21.95%	−14.51%	−13.89%	−15.73%
Max	13.36%	18.51%	33.68%	24.86%	16.03%	16.87%	13.98%	19.46%	7.76%	15.95%	13.13%	10.10%	12.03%
Obs	148	148	148	148	148	148	34	148	148	148	148	148	52
Democratic Presidential Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Mean	1.27%	1.47%	1.41%	1.29%	1.27%	1.29%	0.99%	1.24%	0.86%	0.92%	0.92%	0.87%	0.60%
Standev.S	4.16%	6.28%	6.83%	5.99%	5.50%	6.52%	5.62%	5.69%	3.81%	6.46%	3.99%	4.57%	5.80%
Min	−14.11%	−22.91%	−18.00%	−20.51%	−17.06%	−25.07%	−13.91%	−12.31%	−11.70%	−16.92%	−12.36%	−12.52%	−13.95%
Max	11.48%	24.32%	26.33%	15.89%	19.23%	22.89%	15.07%	19.11%	9.72%	25.12%	10.05%	12.37%	12.47%
Obs	239	165	168	168	168	168	48	168	168	168	168	168	62

reports on the descriptive statistics for the sector-based ETFs included in the study. The summary provides an overview of the return distributions and volatility across sectors, laying the groundwork for subsequent comparative and regression analyses.

As shown in Table 2, most sector ETFs appear to exhibit higher mean returns during Democratic presidential periods compared to Republican periods. To formally test whether these differences are statistically significant, we conduct two-sample t-tests comparing the mean returns of each ETF under Democratic versus Republican administrations. The results of these tests are presented in

Table 3. Two-Sample t Tests Comparing the Mean Returns under Democratic vs. Republican Administration. This table reports the results of two-sample t-tests comparing the mean returns of various ETFs under Democratic versus Republican administrations. The rows labeled R Mean and R Variance present the mean and variance of monthly returns during Republican presidential periods, while D Mean and D Variance report the corresponding statistics during Democratic presidential periods. The column t-stat shows the test statistic from the two-sample t-test, and P (one-tail) and P (two-tail) report the corresponding p-values for the one-tailed and two-tailed significance tests, respectively.

	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
R Mean	0.28%	0.42%	0.28%	0.41%	0.30%	−0.09%	0.96%	0.46%	0.33%	0.63%	0.51%	0.48%	0.72%
R variance	0.21%	0.53%	0.62%	0.53%	0.29%	0.33%	0.37%	0.31%	0.12%	0.33%	0.18%	0.19%	0.19%
D Mean	1.27%	1.47%	1.41%	1.29%	1.27%	1.29%	0.99%	1.24%	0.86%	0.92%	0.92%	0.87%	0.60%
D Variance	0.17%	0.39%	0.47%	0.36%	0.30%	0.43%	0.32%	0.32%	0.15%	0.42%	0.16%	0.21%	0.34%
t Stat	−2.151	−1.356	−1.348	−1.166	−1.569	−1.983	−0.017	−1.235	−1.316	−0.437	−0.868	−0.774	0.123
P (T ≤ t) one-tail	0.016 **	0.088 *	0.089 *	0.122	0.059 *	0.024 **	0.493	0.109	0.095 *	0.331	0.193	0.220	0.451
P (T ≤ t) two-tail	0.032 **	0.176	0.179	0.244	0.118	0.048 **	0.987	0.218	0.189	0.663	0.386	0.440	0.902

***, ** and * represent significance levels at 1%, 5% and 10%, respectively.

.

Table 3 presents the results of the two-sample t-tests. The test statistics are negative for all ETFs except XLRE (SPDR Real Estate), suggesting that average returns during Republican presidential periods tend to be lower than those during Democratic periods across most sectors, with the exception of real estate. Under the one-tailed t-test, the differences are statistically significant for SPY and XLF (Financials) at the 5% level, and for XLE (Energy), XLI (Industrials), and XLP (Consumer Staples) at the 10% level. In the case of the two-tailed t-test, only SPY and XLF exhibit statistically significant differences, both at the 5% significance level.

To further examine the impact of presidential leadership on sector returns, we performed regression analyses. The results are summarized in

Table 4. Empirical Results for Regression Analysis. This table presents the regression analysis results in three specifications. In Specification 1, each ETF’s monthly return is regressed on the Republican dummy variable (Rdummy) which equals 1 during the republican presidential periods and 0 otherwise. In Specification 2, ETF returns are regressed on the market return (SPY) and Rdummy. In Specification 3, ETF returns are regressed on SPY, Rdummy, and the interaction term SPY × RdummySPY. Reported beneath each coefficient are the corresponding p-values.

Panel 1: Regression Analysis of ETFs on Rdummy
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Rdumy	−0.010	−0.010	−0.011	−0.009	−0.010	−0.014	0.000	−0.008	−0.005	−0.003	−0.004	−0.004	0.001
	0.028 **	0.173	0.175	0.239	0.118	0.05 **	0.986	0.219	0.192	0.665	0.384	0.441	0.905
R2	0.012	0.006	0.006	0.004	0.008	0.012	0.000	0.005	0.005	0.001	0.002	0.002	0.000
Adjusted R2	0.010	0.003	0.003	0.001	0.005	0.009	−0.012	0.002	0.002	−0.003	−0.001	−0.001	−0.009
Obs	387	313	316	316	316	316	82	316	316	316	316	316	114
Panel 2: Regression Analysis of ETFs on SPY and Rdummy
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
SPY		1.279	1.021	1.278	1.104	1.153	1.004	1.112	0.523	1.135	0.734	0.488	0.886
		0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***
Rdumy		0.000	−0.003	0.002	−0.001	−0.004	0.002	0.001	−0.001	0.006	0.002	0.000	0.001
		0.962	0.650	0.684	0.801	0.269	0.741	0.695	0.727	0.115	0.503	0.987	0.845
R2		0.706	0.384	0.736	0.804	0.688	0.797	0.769	0.411	0.677	0.627	0.236	0.610
Adjusted R2		0.704	0.380	0.734	0.803	0.686	0.792	0.768	0.407	0.675	0.625	0.231	0.603
Obs		313	316	316	316	316	82	316	316	316	316	316	114
Panel 3: Regression Analysis of ETFs on SPY, Rdummy and Interactive Variable: Rdummy * SPY
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
SPY		1.153	0.909	1.150	1.126	1.220	1.015	1.129	0.523	1.203	0.734	0.452	1.125
		0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***
Rdumy		−0.002	−0.005	0.000	0.000	−0.003	0.002	0.001	−0.001	0.007	0.002	0.000	0.006
		0.650	0.496	0.964	0.886	0.387	0.720	0.645	0.729	0.073	0.510	0.927	0.304
Rdummy * SPY		0.252	0.225	0.256	−0.043	−0.135	−0.021	−0.033	0.001	−0.137	0.001	0.072	−0.465
		0.007 ***	0.127	0.003 ***	0.485	0.129	0.858	0.630	0.991	0.123	0.984	0.474	0.0 ***
R2		0.713	0.389	0.743	0.805	0.690	0.797	0.769	0.411	0.680	0.627	0.237	0.652
Adjusted R2		0.710	0.383	0.741	0.803	0.687	0.789	0.767	0.405	0.676	0.623	0.230	0.642
Obs		313	316	316	316	316	82	316	316	316	316	316	114

***, ** and * represent significance levels at 1%, 5% and 10%, respectively.

.

Panel 1 of the regression analysis shows that the coefficients of the Rdummy variable for various sector ETFs are consistently negative, except for XLRE (SPDR Real Estate). This suggests that Republican presidential periods are generally associated with lower monthly returns across most sectors, with real estate being the only exception. However, the coefficients are statistically significant only for SPY and XLF (SPDR Financials), which aligns with the results of the two-sample t-tests.

Panel 2 incorporates SPY (the market return proxy) as an additional explanatory variable. When SPY is included, the coefficients on the Rdummy variable become minimal and statistically insignificant, while the coefficients on SPY are consistently positive and highly significant across all sector ETFs. This indicates that once overall market performance is accounted for, presidential leadership has no additional explanatory power over sector ETF returns—suggesting that the apparent political effect observed earlier may be largely driven by broader market movements rather than sector-specific political influences.

Panel 3 introduces an interaction term between the market return (SPY) and Rdummy to further examine whether the impact of overall market performance on sector ETFs differs across presidential administrations. The regression results are largely consistent with those in Panel 2: the coefficients for SPY remain statistically significant across all sectors, while the coefficients for Rdummy remain minimal and statistically insignificant.

However, the interaction term (SPY × Rdummy) shows statistical significance only for QQQ and XLRE. Specifically, the coefficient is positive and significant for QQQ, indicating that during Republican presidential periods, the sensitivity of QQQ to overall market returns increases. In contrast, the interaction term is negative and significant for XLRE, suggesting that Republican leadership reduces the influence of broad market movements on the real estate sector. These findings imply that political leadership may modify how certain sectors respond to general market trends, particularly in technology (QQQ) and real estate (XLRE).

5. Empirical Analysis Without Extreme Equity Market Events

One important concern in analyzing the relationship between presidential political affiliation and sector ETF performance is the potential distortion caused by extreme equity market events. Major drawdowns—such as those during financial crises or pandemic-induced selloffs—can significantly influence average return patterns and exaggerate statistical relationships, making it difficult to isolate the true effect of political leadership. To address this issue, we conducted a filtered analysis by removing several notable periods of significant market turmoil from our sample. This refined approach allowed us to re-estimate all analyses under more stable market conditions. By excluding these extreme episodes, we aimed to test the robustness of our findings and assess whether the observed political effects persist once the noise of market crises is removed.

Table 5. Equity Drawdown Events. This table presents detailed information of extreme equity events.

Extreme Equity Drawdown Event	Begin	End	Maximum Drawdown	Months to Drawdown
Post-COVID Inflation	Jan-22	Oct-22	−25%	9
COVID-19	Feb-20	April-20	−34%	1
Global Financial Crisis	Oct-07	Mar-09	−57%	17
Tech Bubble	Mar-00	Oct-02	−49%	31

presents the major equity market drawdown events in recent history. To perform our robustness check, we exclude these periods from the sample and re-conduct the full set of analyses.

Table 6. Descriptive Statistics for ETF Returns (Excluding Extreme Equity Events). This table presents summary statistics for monthly returns of sector-specific exchange-traded funds (ETFs), excluding dates associated with extreme equity market events. Returns are expressed in percentage terms.

All Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Data History	01/93–03/25	03/99–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	06/18–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	12/98–03/25	11/15–03/25
Mean	1.43%	2.00%	1.14%	1.81%	1.45%	1.31%	1.87%	1.43%	0.86%	1.26%	1.04%	1.13%	1.14%
Standev.S	3.63%	4.97%	6.11%	4.87%	4.62%	5.13%	4.88%	4.76%	3.28%	5.51%	3.57%	3.90%	4.46%
Min	−14.11%	−12.09%	−16.00%	−13.44%	−11.22%	−11.77%	−8.28%	−12.31%	−11.70%	−16.92%	−9.76%	−12.17%	−9.63%
Max	11.48%	24.32%	27.97%	15.89%	19.23%	22.89%	15.07%	19.11%	9.70%	25.12%	9.28%	10.44%	12.47%
Obs	324	250	253	253	253	253	69	253	253	253	253	253	101
Republican Presidential Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Mean	1.19%	1.69%	1.21%	1.56%	1.37%	1.09%	1.21%	1.09%	0.88%	1.44%	0.88%	1.50%	1.00%
Standev.S	3.34%	4.73%	6.29%	4.84%	4.09%	4.20%	5.25%	4.25%	2.80%	4.42%	3.41%	3.34%	3.36%
Min	−9.34%	−12.09%	−16.00%	−13.44%	−11.22%	−11.68%	−8.28%	−9.83%	−9.75%	−8.76%	−9.76%	−6.94%	−8.41%
Max	10.88%	12.93%	27.97%	15.84%	16.03%	16.87%	12.09%	11.85%	7.76%	12.38%	8.09%	10.10%	12.03%
Obs	108	108	108	108	108	108	31	108	108	108	108	108	49
Democratic Presidential Periods
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Mean	1.55%	2.29%	1.08%	2.01%	1.51%	1.47%	2.40%	1.68%	0.85%	1.13%	1.15%	0.85%	1.26%
Standev.S	3.77%	5.15%	6.00%	4.89%	5.00%	5.73%	4.56%	5.10%	3.60%	6.21%	3.69%	4.26%	5.32%
Min	−14.11%	−9.23%	−14.76%	−9.93%	−9.78%	−11.77%	−6.83%	−12.31%	−11.70%	−16.92%	−7.97%	−12.17%	−9.63%
Max	11.48%	24.32%	22.44%	15.89%	19.23%	22.89%	15.07%	19.11%	9.70%	25.12%	9.28%	10.44%	12.47%
Obs	216	142	145	145	145	145	38	145	145	145	145	145	52

shows the descriptive statistics for the ETF returns during the periods when extreme market events are excluded.

A two-sample t test is then performed to study the statistical difference in the average mean returns for various ETFs between Republican presidential periods vs. Democratic presidential periods, and the results are shown in

Table 7. Two-Sample t-Test of ETF Returns (Excluding Extreme Market Events). This table reports the results of two-sample t-tests comparing average monthly returns of sector-specific ETFs across two sample periods (e.g., Democratic vs. Republican presidencies) when the extreme market events are excluded. The rows labeled R Mean and R Variance present the mean and variance of monthly returns during Republican presidential periods, while D Mean and D Variance report the corresponding statistics during Democratic presidential periods. The column t-stat shows the test statistic from the two-sample t-test, and P (one-tail) and P (two-tail) report the corresponding p-values for the one-tailed and two-tailed significance tests, respectively.

	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
R Mean	1.19%	1.63%	1.21%	1.56%	1.37%	1.09%	1.21%	1.09%	0.88%	1.44%	0.88%	1.50%	1.00%
R variance	0.11%	0.22%	0.40%	0.23%	1.51%	0.18%	0.28%	0.18%	0.08%	0.20%	0.12%	0.11%	0.11%
D Mean	1.55%	2.28%	1.08%	2.01%	0.17%	1.47%	2.40%	1.68%	0.85%	1.13%	1.15%	0.85%	1.26%
D Variance	0.14%	0.27%	0.36%	0.24%	0.25%	0.33%	0.21%	0.26%	0.13%	0.39%	0.14%	0.18%	0.28%
t Stat	−0.877	−1.037	0.166	−0.724	−0.233	−0.600	−0.994	−1.006	0.080	0.458	−0.604	1.352	−0.298
P (T ≤ t) one-tail	0.191	0.150	0.434	0.235	0.408	0.275	0.162	0.158	0.468	0.324	0.273	0.089	0.383
P (T ≤ t) two-tail	0.381	0.301	0.868	0.470	0.816	0.549	0.324	0.316	0.937	0.647	0.546	0.178	0.766

***, ** and * present significance levels at 1%, 5% and 10%, respectively.

.

In contrast to the findings based on the full sample, the results from the two-sample t-tests—conducted after excluding periods of extreme equity market drawdowns—reveal a notable shift in the direction of the test statistics. Specifically, the t-statistics for XIRE turned negative, while the t-statistics for XLE, XLP, XLB, and XLU became positive. This suggests that, during more stable market conditions, these sector ETFs exhibited higher average returns under Republican presidential periods compared to Democratic ones. However, it is important to note that none of these t-statistics reached conventional levels of statistical significance. As such, the differences in mean returns across political administrations should be interpreted with caution, as they may not reflect meaningful or robust relationships.

Regression analysis yields similar findings. As presented in

Table 8. Regression Results for ETF Returns (Excluding Extreme Equity Market Events). In this table, we present regression analysis results for samples excluding extreme equity events. In Specification 1, each ETF’s monthly return is regressed on the Republican dummy variable (Rdummy) which equals 1 during the republican presidential periods and 0 otherwise. In Specification 2, ETF returns are regressed on the market return (SPY) and Rdummy. In Specification 3, ETF returns are regressed on SPY, Rdummy, and the interaction term SPY × RdummySPY. Reported below each coefficient are the corresponding p-values.

Panel 1: Regression Analysis of ETFs on Rdummy
	SPY	QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
Rdumy	−0.004	−0.007	0.001	−0.004	−0.001	−0.004	−0.012	−0.006	0.000	0.003	−0.003	0.006	−0.003
	0.400	0.306	0.867	0.470	0.821	0.566	0.317	0.328	0.939	0.663	0.551	0.193	0.769
R2	0.002	0.004	0.000	0.002	0.000	0.001	0.015	0.004	0.000	0.001	0.001	0.007	0.001
Adjusted R2	−0.001	0.000	−0.004	−0.002	−0.004	−0.003	0.000	0.000	−0.004	−0.003	−0.003	0.003	−0.009
Obs	324	250	253	253	253	253	69	253	253	253	253	253	101
Panel 2: Regression Analysis of ETFs on SPY and Rdummy
		QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
SPY		1.147	1.003	1.139	1.127	1.179	1.017	1.168	0.586	1.251	0.738	0.392	0.838
		0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***
Rdumy		−0.003	0.004	−0.001	0.002	0.000	−0.005	−0.003	0.002	0.007	−0.001	0.008	0.000
		0.358	0.512	0.710	0.516	0.912	0.352	0.359	0.537	0.106	0.839	0.103	0.999
R2		0.690	0.347	0.707	0.765	0.682	0.764	0.781	0.412	0.665	0.552	0.137	0.487
Adjusted R2		0.688	0.342	0.704	0.764	0.680	0.757	0.779	0.408	0.663	0.549	0.130	0.476
Obs		250	253	253	253	253	69	253	253	253	253	253	101
Panel 3: Regression Analysis of ETFs on SPY, Rdummy and Interactive Variable Rdummy * SPY
		QQQ	XLE	XLK	XLI	XLF	XLC	XLY	XLP	XLB	XLV	XLU	XIRE
SPY		1.076	0.973	1.056	1.149	1.235	1.024	1.183	0.610	1.355	0.741	0.440	1.274
		0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***	0.0 ***
Rdumy		−0.006	0.003	−0.004	0.003	0.002	−0.005	−0.002	0.003	0.010 **	−0.001	0.009 *	0.011 *
		0.130	0.645	0.245	0.389	0.689	0.410	0.490	0.411	0.018	0.875	0.062	0.069
Rdummy * SPY		0.190	0.080	0.224	−0.060	−0.154	−0.013	−0.040	−0.065	−0.283	−0.008	−0.131	−0.783
		0.061 *	0.659	0.020 **	0.465	0.146	0.928	0.626	0.482	0.015 **	0.924	0.324	0.0 ***
R2		0.694	0.347	0.713	0.766	0.685	0.764	0.781	0.414	0.673	0.552	0.141	0.591
Adjusted R2		0.691	0.339	0.709	0.763	0.681	0.753	0.778	0.407	0.669	0.547	0.130	0.579
Obs		250	253	253	253	253	69	253	253	253	253	253	101

***, ** and * present significance levels at 1%, 5% and 10%, respectively.

, Panel 1, when the returns of various ETFs are regressed on the Rdummy variable, none of the coefficients are statistically significant. This suggests that presidential political affiliation does not have a meaningful or consistent impact on sector-specific ETF performance during the periods analyzed. Consistent with the results from the two-sample t-tests, the coefficients on Rdummy are negative for most ETFs, with the exceptions being XLI (Industrials), XLF (Financials), XLP (Consumer Staples), XLB (Materials), XLU (Utilities), and XLRE (Real Estate), where the coefficients are positive—though still not statistically significant.

Panel 2 demonstrates that the coefficients for SPY are positive and statistically significant at the 1% level across all sector ETFs, underscoring the strong influence of overall market performance on sector-specific returns. Building on this, Panel 3 reveals that the interaction terms between SPY and the Republican dummy variable (Rdummy) are statistically significant for QQQ, XLK (Technology), XLB (Materials), and XIRE (Real Estate), suggesting that presidential leadership may alter how sector returns respond to market movements.

Notably, the direction of these interaction effects varies by sector. For QQQ and XLK, the positive interaction coefficients indicate that Republican presidential periods tend to amplify the sensitivity of technology-related ETFs to broader market movements. In contrast, the negative interaction terms observed for XLB and XIRE suggest that Republican leadership dampens the responsiveness of the materials and real estate sectors to market-wide fluctuations.

Additionally, the coefficients for the Rdummy variable are positive and statistically significant for XLB (Materials), XLU (Utilities), and XIRE (Real Estate), although the magnitudes of these coefficients are relatively small. This suggests that, beyond the broader market effects, Republican administrations are associated with modest improvements in the performance of these sectors.

6. Conclusions

This study began with a simple but important question: does presidential political affiliation influence the performance of sector-specific exchange-traded funds (ETFs)? It is a question rooted in both popular perception and academic inquiry—many investors and analysts believe that markets respond differently depending on which party controls the White House. But do these political shifts truly move the needle in meaningful ways for investors? And if so, is that influence consistent across sectors?

To answer these questions, we adopted a two-pronged analytical approach. First, we looked at the full historical record, including major market disruptions when politics and economics collide most visibly. Then, to isolate political effects from broader market volatility, we repeated our analysis with those extreme drawdown periods removed.

What we found was both nuanced and revealing. In the full-sample analysis, there was indeed some evidence suggesting that presidential affiliation matters—at least on the surface. Most sector ETFs, with the exception of real estate (XLRE), showed lower average returns during Republican presidencies compared to Democratic ones. These differences were statistically significant for a few sectors, especially when we focused on the broader market (SPY) and financials (XLF). These findings echoed earlier research, including the widely cited work of Santa-Clara and Valkanov (2003), suggesting that political cycles can leave a measurable imprint on market performance.

Digging deeper, we used regression models to control broader market movements. In the first model, Republican administrations were associated with negative returns for most ETFs, and again, SPY and XLF stood out. But once we accounted for overall market performance by including SPY as a control variable, the story changed. The influence of political affiliation all but disappeared, while SPY itself emerged as the dominant force explaining ETF performance. This shift suggests that the apparent political effect might actually be a reflection of how the broader market behaved during those administrations, not a direct consequence of party control.

Still, we were not quite ready to close the book. We wondered whether political leadership might not change average returns outright but instead shape how different sectors respond to the market. To explore this, we introduced an interaction between SPY and our political variable. Interestingly, technology (QQQ) became more responsive to market movements during Republican presidencies, while real estate (XLRE) seemed to become less sensitive. These subtle but significant patterns hinted at a more complex relationship between politics and markets—one that goes beyond simple averages.

We then turned to our second approach, removing periods of extreme market stress. The findings revealed a notable shift: once these crisis episodes were excluded, the differences in average returns across political regimes largely vanished. This suggests that, absent the emotional volatility of market extremes, political leadership no longer appeared to exert a meaningful influence on sector performance.

The regression results told a similar story. Across all models, the presidential dummy variable lost its explanatory power. However, the interactions between politics and market sensitivity remained. Technology ETFs, for instance, still reacted more strongly to market movements during Republican periods, while sectors like materials and real estate became less reactive. These patterns suggest that while political affiliation might not directly drive sector performance, it does appear to modulate how sensitive different sectors are to the broader market environment.

Taken together, our results suggest that the observed political differences in sector ETF returns are largely driven by periods of extreme market turbulence. Remove those moments, and the direct effect of political leadership mostly vanishes. But even under more stable conditions, the way certain sectors respond to market forces can still depend on who is in office. It is not that politics dictates performance, but that it subtly changes the rules of engagement between sectors and the market at large.

Our study contributes to literature in several important ways. First, by focusing on sector-specific ETFs rather than broad indices, we offer a more detailed view of political influence—one that highlights the diversity of sectoral responses to the same political environment. Second, we bridge two areas of research that have often evolved in parallel: the study of political cycles and the analysis of ETF behavior. In doing so, we propose ETFs as a valuable tool for both researchers and investors interested in political risk.

Third, our two-stage empirical design—before and after controlling for market crises—offers a clearer and more credible account of when and how politics matters in the market. Many past studies may have overstated political effects by failing to account for these major distortions. Fourth, our results carry practical implications. For investors looking to navigate politically charged environments, knowing how different sectors behave under various administrations can inform smarter, more resilient portfolio choices.

Finally, our work opens the door to new avenues of inquiry. Future research could take these questions to other asset classes, or to the industry or firm level or extend the analysis to other countries using international ETFs. One direction would be to extend the analysis by incorporating additional asset classes, such as bonds, commodities, or international equities, to assess whether partisan effects extend beyond sector ETFs. Another fruitful extension would be to investigate potential non-linear dynamics or regime-switching behavior, capturing the possibility that political influences vary across different phases of the business cycle or in response to major economic shocks. Finally, applying machine learning or other advanced econometric techniques could uncover more subtle patterns that traditional approaches may overlook. Such efforts would further enrich the understanding of how political leadership interacts with financial markets.

It is important to note that our sample period spans several decades, during which U.S. financial markets have experienced significant structural shifts. Major events such as the technology boom of the late 1990s, increased globalization of capital markets in the 2000s, the Global Financial Crisis of 2008, and the subsequent regulatory reforms fundamentally altered the dynamics of asset pricing and sector performance. These events may represent structural breaks that could affect return behavior independently of presidential partisanship. For instance, post-2008 regulatory changes had a lasting impact on the financial sector, while the tech boom disproportionately shaped returns in the technology sector. Although our primary focus is on partisan influences, it is possible that these regime shifts interact with or even overshadow partisan effects in certain periods. Future research could formally test for structural breaks or conduct sub-sample robustness analyses to further disentangle partisan effects from broader market transformations.

In summary, our study tells a story of complexity. Political leadership does influence the market—but not always in the ways we expect. Sometimes, the biggest impact is not on the numbers themselves, but on how those numbers respond to the tides of the broader economy. By teasing apart these relationships, we hope to offer both theoretical insight and practical guidance to anyone navigating the ever-changing intersection of politics and markets.