Regulatory and Political Risk, and Multinational Corporate Strategy in Kenya

Introduction

Multinational corporations (MNCs) are central to Kenya’s economic development. They contribute significantly to foreign direct investment (FDI), employment creation, and technology transfer, positioning Kenya as a regional investment hub in East Africa. However, the country’s business environment is shaped by fluctuating political dynamics, evolving regulatory frameworks, and periodic governance challenges that influence both the scale and nature of multinational investment. Episodes of electoral tension, policy uncertainty, and changing taxation or licensing rules have created a setting where political and regulatory risk can substantially affect corporate strategy and investment behaviour.

Political and regulatory risks are widely recognized as critical determinants of corporate performance in developing economies. For MNCs, such risks manifest through uncertain government policies, corruption, contract enforcement weaknesses, or shifts in trade and taxation regimes. These factors can increase operational costs, delay investment decisions, and trigger corporate adaptations such as reduced capital expenditure, increased cash reserves, or strategic partnerships with local firms. Understanding how these risks shape MNC behaviour in Kenya is therefore vital for both policymakers and business leaders seeking to foster stable investment climates and sustainable growth.

This study examines how political and regulatory risk influences the strategic and investment behaviour of multinational corporations operating in Kenya between 2010 and 2024. Using secondary data drawn from company financial reports, the Central Bank of Kenya, the World Bank’s Worldwide Governance Indicators, and UNCTAD’s FDI database, the paper empirically evaluates the relationship between shifts in Kenya’s governance environment and MNC investment decisions.

The theoretical link between political/regulatory risk and corporate behaviour draws on institutional theory and the resource-dependence perspective. Institutional theory (DiMaggio & Powell, 1983) posits that organizations adapt to the institutional constraints of their environments to secure legitimacy and survival. Accordingly, MNCs in volatile political contexts may adopt conservative financial policies, enhance CSR visibility, or localize operations to align with governmental expectations. Resource-dependence theory (Pfeffer & Salancik, 1978) further suggests that firms dependent on external actors such as governments for licenses or contracts must manage those dependencies strategically.Based on these perspectives, the study formulates three hypotheses: 1) H1: Periods of increased political and regulatory risk in Kenya are associated with lower MNC investment and capital expenditure. 2) H2: MNCs increase liquidity and reduce leverage during high-risk periods to buffer against uncertainty. 3) H3: The effect of political and regulatory risk is stronger in sectors highly exposed to government regulation or procurement.

The study contributes to the literature in three key ways.First, it provides firm-level evidence from a developing economy where political risk remains an underexplored determinant of business performance. Second, it integrates governance indicators with corporate financial metrics to capture the multi-dimensional nature of risk. Third, it generates practical insights for policy and business strategy highlighting how predictable regulatory regimes can sustain investment even under political uncertainty.The remainder of the paper is structured as follows. Section 2 reviews relevant literature on political risk, multinational investment, and corporate strategy. Section 3 outlines the data sources and methodological approach. Section 4 presents the results, while Section 5 discusses their implications for theory and practice. Section 6 concludes, and Section 7 highlights study limitations and avenues for future research.

Literature Review

2.1. Multinational Corporations and Political Risk

The international business literature has long established that multinational corporations (MNCs) face distinctive risks when investing in developing economies. Rooted in the classical work of Hymer (1976) and Dunning’s eclectic paradigm (1980), foreign direct investment (FDI) is conditioned not only by market opportunities but also by host-country institutional quality. Political risk defined as the probability that political decisions, events, or conditions will negatively affect corporate profitability remains a decisive factor shaping MNC entry and operation strategies. Kobrin (1979) emphasized that such risks arise from host-government actions including expropriation, policy reversals, and trade restrictions. Later studies, such as Henisz (2000), empirically demonstrated that countries with stronger political constraints and credible institutions attract greater levels of FDI because investors anticipate policy stability. Conversely, in environments characterized by corruption, weak rule of law, and frequent regulatory changes, MNCs adopt risk-averse postures through delayed investments or smaller capital commitments (Busse & Hefeker, 2007).

In sub-Saharan Africa, political and institutional factors have repeatedly been shown to moderate foreign investment. Asiedu (2006) finds that political stability, quality of bureaucracy, and corruption control strongly determine Africa’s ability to attract FDI. Similarly, Morisset (2000) argues that regulatory transparency and government credibility matter more than tax incentives. Kenya, as a regional hub, reflects these dynamics: despite its relatively diversified economy, investment inflows fluctuate with electoral cycles. This suggests the sensitivity of MNCs to political uncertainty and regulatory inconsistency.

2.2. Regulatory Risk and Corporate Strategy

Beyond formal political stability, MNCs must contend with regulatory risk the unpredictability of policy interpretation, enforcement, and reform. Regulatory risk encompasses changes in tax rules, licensing procedures, sectoral regulations, and labour or environmental standards (Bergara, Henisz & Spiller, 1998). These uncertainties influence how firms plan long-term investments, allocate capital, and manage compliance costs. The real options theory (Dixit & Pindyck, 1994) posits that when investment outcomes are irreversible and uncertainty is high, firms defer or stage investments until political or regulatory conditions clarify.

Empirical evidence supports this behaviour. Julio and Yook (2012) show that firms globally reduce investment expenditures during election periods due to policy uncertainty. Similarly, Bloom et al. (2007) demonstrate that heightened uncertainty leads to lower employment and capital expenditure, as firms value flexibility over commitment. MNCs also respond strategically by adopting hedging tactics: building alliances with domestic partners, engaging in lobbying, or expanding corporate social responsibility (CSR) initiatives to secure political legitimacy (Frynas & Mellahi, 2003). In the African context, Kolstad and Wiig (2012) highlight that MNCs in resource-rich countries often use CSR and community engagement to mitigate political opposition and maintain operating licenses.

In Kenya, the regulatory environment has evolved considerably since the early 2000s. Reforms in the telecommunications, financial, and energy sectors increased foreign participation, yet recurring policy shifts such as changes in tax laws, localization requirements, and environmental approvals introduce uncertainty. The Kenyan experience aligns with the argument that institutional volatility constrains corporate planning and increases transaction costs (North, 1990). Firms facing such environments adopt defensive strategies, such as reducing capital intensity, increasing liquidity reserves, or diversifying geographically across neighbouring East African markets.

2.3. Political Economy of MNCs in Developing Economies

From a political economy perspective, MNCs are not passive actors but influential participants shaping host-country policy. Strange (1996) and Stopford & Strange (1991) describe the “triangular diplomacy” among states, firms, and international institutions, where MNCs use bargaining power to negotiate favourable terms. In developing states, the asymmetry between capital-rich corporations and resource-dependent governments often grants MNCs leverage to influence policy outcomes. Yet, this leverage is double-edged: excessive political exposure can invite populist backlash or policy reversal, especially during elections or leadership transitions (Jensen, 2003).

Empirical work in Africa reveals mixed evidence. Osei-Kufuor and Dorian (2021) find that while MNCs can secure short-term regulatory advantages through lobbying, such advantages erode when government legitimacy declines. Similarly, Darley (2012) argues that in African extractive industries, MNCs’ engagement with ruling elites sometimes amplifies governance risks rather than mitigating them. For Kenya, case studies of multinational banks, telecom firms, and agribusinesses suggest that corporate strategies often blend compliance with informal relationship management, reflecting both the strength and limits of local institutions (Kamau & Munene, 2020).

2.4. Empirical Evidence from Kenya and East Africa

Existing empirical analyses of FDI in Kenya identify multiple determinants beyond political risk. Mwega and Ngugi (2006) attribute FDI fluctuations to macroeconomic reforms, infrastructure development, and market size, but they note that political instability during election years consistently undermines investor confidence. World Bank data show that FDI inflows dipped sharply around the 2007–2008 post-election violence, then recovered during periods of relative stability. Moreover, Transparency International’s corruption perception index indicates persistent governance challenges, while the World Governance Indicators highlight stagnation in regulatory quality and control of corruption since 2015.

Sector-specific studies provide deeper insight. In telecommunications, global firms such as Vodafone and Airtel have thrived under relatively predictable regulation, whereas energy and extractive sectors remain constrained by licensing opacity and policy reversals (Kinyua, 2019). Manufacturing MNCs report high compliance costs linked to customs and taxation unpredictability (KNBS, 2023). These findings collectively demonstrate that Kenya’s political-regulatory environment remains a critical variable in MNC performance and strategy. Yet, few quantitative studies systematically link governance indicators to firm-level financial data a gap this paper seeks to address.

Methods

Research Design

This paper adopts a longitudinal analytical design using secondary quantitative data to investigate the influence of political and regulatory risk on the investment and financial strategies of multinational corporations (MNCs) operating in Kenya. The design integrates firm-level panel data with macro-level governance and risk indicators to capture both temporal and structural variations between 2010 and 2024. Because the variables of interest evolve over time, the panel framework enables control for unobservable firm-specific characteristics and national economic shocks that could otherwise bias cross-sectional results. The study’s analytical logic is associative rather than experimental, seeking to estimate the strength and direction of statistical relationships rather than establish direct causality. Nevertheless, the use of firm fixed effects, lagged variables, and difference-in-differences estimation allows for a quasi-causal interpretation consistent with established political economy literature.

The research is grounded on the theoretical assumptions of institutional theory and political risk theory, which together posit that firm strategy is conditioned by the quality and predictability of the institutional environment. Political instability, regulatory unpredictability, and policy inconsistency increase perceived risk, leading MNCs to adjust investment, financing, and liquidity policies accordingly. Therefore, the analytical framework is structured to capture how variations in political and regulatory indicators translate into quantifiable changes in capital expenditure, leverage, cash holdings, and profitability across time and firms.

Sample and Population

The analytical population for this study consists of multinational corporations (MNCs) with active operations in Kenya between 2010 and 2024. An MNC is defined as any firm with majority foreign ownership or a foreign parent company that exerts managerial control over its Kenyan subsidiary. The population covers firms across the telecommunications, energy, manufacturing, financial services and agribusiness sectors. From this population a purposive sample was drawn based on three selection criteria: (i) availability of audited annual reports or publicly-filed financial statements for the Kenyan affiliate over the study period, (ii) continuous operation in Kenya (no major interruptions) within 2010-2024, and (iii) disclosure of sufficient financial data (balance sheet, income statement items) to compute key variables. Applying these criteria yielded a total of n = 72 multinational affiliates, producing an unbalanced panel of approximately 950 firm-year observations (2010-2024 inclusive). The full listing of sampled firms, domiciles, sectors, and observation counts is provided in Appendix B.

Samples of firm-specific data were obtained from the audited annual reports of the Kenyan affiliates, supplemented by filings submitted to the Nairobi Securities Exchange (NSE, 2025) where applicable, and complemented with the Orbis Global Database (Bureau van Dijk, 2025) database, downloaded on 10 May 2025. Macroeconomic and institutional indicators were drawn from publicly available secondary sources, including: the Worldwide Governance Indicators (WGI) from the World Bank (data through 2023) (World Bank 2025); the Armed Conflict Location & Event Data (ACLED, 2025) Kenya country file (downloaded March 2025); and statistical publications of the Central Bank of Kenya (CBK) and the Kenya National Bureau of Statistics (KNBS). Each source was selected for its methodological transparency, longitudinal coverage and internationally‐recognised reliability. Because the study relies exclusively on secondary data, no direct engagement with human participants was required and thus formal ethical approval was not applicable.

Data Sources and Harmonisation

The dataset merges multiple streams of secondary data to ensure comprehensive coverage of both firm-level financial performance and macro-institutional conditions. Firm‐level financial statement variables total assets, capital expenditure, cash holdings, total liabilities, and operating income – were collected in Kenyan shillings (KES) and subsequently converted into constant 2024 United States dollars (USD) to remove inflation and currency‐unit distortions. The conversion employed the following formula:

$${\mathrm{Value}}_{\mathrm{USD},t}=\frac{{\mathrm{Value}}_{\mathrm{KES},t}}{E{R}_t}\times \frac{\mathrm{CP}{I}_{2024}}{\mathrm{CP}{I}_t},$$

where $E{R}_t$ represents the average exchange rate (KES per USD) in year $t$, and $\mathrm{CP}{I}_t$ is the consumer price index for year $t$. This transformation standardizes all financial data to comparable real terms.

Political and regulatory indicators were matched to corresponding fiscal years. The WGI database provides annual measures of Regulatory Quality (RQ), Rule of Low (RL), and Government Effectiveness (GE), each scaled from -2.5 to +2.5. These indices were converted into standardized $z$-scores to ensure comparability across indicators, using:

$$Z_{\mathrm{it}}=\frac{X_{\mathrm{it}}-\bar{X}}{{\sigma }_X},$$

where $X_{\mathrm{it}}$ is the raw governance score for indicator $i$ in year $t_r\bar{X}$ is its mean, and ${\sigma }_X$ is the standard deviation across all years.

Event-level political instability was proxied by the logarithm of the annual number of protests, violent demonstrations, or politically motivated conflicts recorded in the ACLED dataset (ACLED, 2025). To smooth extreme variation, the transformation used was:

$${\mathrm{Conflict}}_t=\log \left(1+{\mathrm{Events}}_t\right),$$

In order to reduce extreme skewness in event counts. A binary variable ElectionYear was coded equal to 1 in Kenya’s national general election years (2013, 2017, 2022) and 0 otherwise. Sector‐specific regulatory shocks were identified through manual content analysis of Kenya Gazette notices (2010–2024) and coded as RegulatoryShock_j,t =1 for years in which a major licensing or regulatory change affected industry j; 0 otherwise (see Appendix C).

Variable Construction

The dependent variables were designed to capture MNCs’ financial and investment responses to variations in political and regulatory conditions. Five firm-level indicators were computed directly from financial statements:

Capital Expenditure Ratio (CapEx_Ratio) — defined as ${\mathrm{CapEx}}_t/{\mathrm{TotalAssets}}_{t-1}$, indicating the intensity of investment relative to firm size.

Investment Growth (InvGrowth) - measured as (TotalAssets ${}_t-$ TotalAssets $\left({}_{t-1}\right)/$ Total Assets ${}_{t-1}$, representing annual expansion in the firm's asset base.

Cash Ratio (CashRatio) - computed as CashAndEquivalents ${}_t/$ TotalAssets ${}_t$, reflecting liquidity preferences under risk.

Leverage (Leverage) - expressed as TotalLiabilities ${}_t/$ TotalAssets ${}_t$, capturing debt dependence.

Return on Assets (ROA) - calculated as Operating Income ${}_t/$ TotalAssets ${}_{t-1}$, indicating profitability.

All variables were winsorized at the 1st and 99th percentiles to minimize outlier influence. The resulting dataset was tested for stationarity using panel unit root tests, confirming that all series were weakly stationary after demeaning.

To aggregate the multiple governance and political indicators into a unified risk measure, a Political and Regulatory Risk Index (PRRI) was constructed. For each year $t$ and indicator $i$, the normalized score was computed as:

$$x_{i,t}^{\text{'}}=\frac{x_{i,t}-\min \left(x_i\right)}{\max \left(x_i\right)-\min \left(x_i\right)},$$

which rescales the raw data into the interval $\left[\mathrm{0,1}\right]$. Directional weights $w_i$ were then assigned to align the directionality of the variables: $w_i=+1$ if higher values indicated stability (e.g., Regulatory Quality, Rule of Law), and $w_i=-1$ if higher values represented greater instability (e.g., Conflict Intensity). The weighted component scores were then aggregated according to:

$$\mathrm{PRR}{I}_t=\sum _{i=1}^k{w}_i{x}_{i,t}^{\text{'}}$$

A practical illustration clarifies this procedure. Suppose the maximum and minimum annual counts of conflict events during the study period were 600 and 40 respectively. If in 2017 there were 400 events, the normalized conflict score would be $(400-40)/(600-40)=0.643$. Applying a negative weight $(w=$ -1 ) yields a weighted score of -0.643 for that year. If the standardized Regulatory Quality index in 2017 was 0.35 , with $w=+1$, its contribution becomes +0.35 . Summing across all weighted components produces the PRRI for 2017. Higher PRRI values (less negative) thus represent a more stable political-regulatory environment.

Control variables were incorporated to isolate the effects of political risk from firm-specific characteristics. Firm size ( Size ${}_{\mathrm{it}}$ ) was measured as the natural logarithm of total assets; Firm age ( Age ${}_{\mathrm{it}}$ ) was computed as years since incorporation; Foreign ownership share $\left(\right)$ Foreign $\left({}_{\mathrm{it}}\right)$ represented the proportion of equity held by non-Kenyan investors; and Industry dummies controlled for sectoral fixed effects. At the macro level, GDP growth, inflation, and exchange rate volatility were included as continuous variables sourced from CBK and World Bank (2025) databases (CBK, 2025). These controls mitigate omitted-variable bias by capturing broader economic trends influencing investment decisions.For full replication, raw data filenames, transformation code scripts (in R version 4.3.1 and Stata 17), and the complete variable dictionary are provided in Appendix A.

Model Specification

The baseline empirical relationship was estimated using a fixed-effects (FE) panel model to control for unobserved heterogeneity across firms and time. The model is expressed as:

$$Y_{\mathrm{it}}={\alpha }_i+{\lambda }_t+{\beta }_1\mathrm{PRR}{I}_t+{\beta }_2{X}_{\mathrm{it}}+{\epsilon }_{\mathrm{it}},$$

where $Y_{\mathrm{it}}$ denotes one of the five dependent variables for firm $i$ in year $t;{\alpha }_i$ captures firm-specific effects; ${\lambda }_t$ represents time effects common to all firms; $\mathrm{PRR}{I}_t$ is the composite political-regulatory risk index; $X_{\mathrm{it}}$ is the vector of control variables; and ${\epsilon }_{\mathrm{it}}$ is the idiosyncratic error term. The coefficients ${\beta }_1$ and ${\beta }_2$ estimate the marginal effects of risk and control factors, respectively. Standard errors were clustered at the firm level to correct for serial correlation and heteroskedasticity.

Recognizing that firm financial behavior exhibits temporal persistence, a dynamic panel model was estimated to incorporate lagged dependent variables, expressed as:

$$Y_{\mathrm{it}}={\alpha }_i+{\lambda }_t+\rho Y_{i,t-1}+{\beta }_1\mathrm{PRR}{I}_{t-1}+{\beta }_2{X}_{\mathrm{it}}+u_{\mathrm{it}}$$

Here, $\rho$ measures the degree of adjustment inertia. Because the lagged term introduces potential endogeneity, estimation relied on the System Generalized Method of Moments (GMM) technique (Arellano & Bover, 1995; Blundell & Bond, 1998). The validity of instruments was confirmed using the Hansen $J$-test, while the Arellano-Bond $\mathrm{AR}\left(2\right)$ statistic ensured the absence of second-order serial correlation.

For regulatory reforms affecting only certain industries, the study employed a difference-in-differences (DiD) estimator to isolate causal effects of specific regulatory changes. The model takes the form:

$$Y_{\mathrm{it}}={\alpha }_i+{\lambda }_t+\delta \left({\mathrm{Treated}}_i\times {\mathrm{Post}}_t\right)+\gamma X_{\mathrm{it}}+{\epsilon }_{\mathrm{it}},$$

where Treated ${}_i$ equals 1 for firms in affected sectors and ${\mathrm{Post}}_t$ equals 1 for years after the reform. The coefficient $\delta$ captures the average treatment effect of the regulatory shock on the treated group. Pretreatment parallel trends were verified through graphical and statistical tests based on event-study estimations of leads and lags.

Endogeneity concerns were further addressed through an instrumental variable (IV) framework using exogenous regional political shocks as instruments. Specifically, the number of political unrest events in neighboring countries (Uganda, Tanzania, and Ethiopia) served as instruments for domestic instability, under the assumption that cross-border unrest affects Kenya's perceived political risk but not individual firm performance directly. The first-stage regression tested instrument strength via the F-statistic, while the second-stage regression estimated the structural parameters.

Data Treatment and Robustness Procedures

Data quality was ensured through extensive cleaning and harmonization. Missing observations below 10 percent per variable were addressed through listwise deletion, while higher missingness was resolved using multiple imputation by chained equations (MICE) under the assumption of data missing at random. Outliers were winsorized at the 1st and 99th percentiles, and robustness checks using 5th–95th percentile trimming confirmed the stability of the results. All statistical computations were performed using Stata 17 and R software (packages fixest, plm, ivreg).

Several robustness tests were implemented. First, alternative specifications of the PRRI index were estimated using principal component analysis (PCA) instead of weighted summation to confirm consistency of results. Second, models were re-estimated with lagged governance indicators (PRRI_t−1) to account for delayed corporate responses to institutional changes. Third, the sample was stratified by sector to test whether the risk–investment relationship varied across industries with differing regulatory exposure. Lastly, placebo tests were conducted by assigning pseudo-reform dates to confirm that detected effects were not driven by random temporal coincidence.

Findings

This section presents the results derived from the analysis of multinational corporations (MNCs) operating in Kenya from 2010 to 2024. The findings are organized into two main parts: descriptive findings, which summarize the data and examine preliminary relationships, and empirical findings, which test the hypothesized effects of political and regulatory stability on MNC financial performance and investment behaviour. Together, these results provide quantitative evidence on how political-regulatory dynamics shape corporate financial decisions in Kenya’s investment environment.

4.1 Descriptive Findings

This subsection provides a summary of the main variables used in the analysis, offering a preliminary overview of patterns and relationships before moving to regression estimations. It includes measures of firm investment intensity, profitability, leverage, liquidity, and the Political and Regulatory Risk Index (PRRI), which captures the stability of Kenya’s institutional and policy environment.Table 4.1 presents the descriptive statistics of all variables across 72 multinational firms and 950 firm-year observations. These statistics provide insight into the central tendencies and variation in firm performance indicators.

Note. All financial variables are measured in constant 2024 USD. Source: Author’s computation based on secondary data (NSE (2025), Orbis, World Bank (2024-2025), CBK, 2025).

As Table 4.1 indicates, the average capital expenditure ratio (0.18) and investment growth (0.126) suggest moderate but sustained reinvestment among MNCs. The average political and regulatory risk index (0.03) shows slight positive stability, with a broad range (−0.58 to 0.64), reflecting episodes of both uncertainty and reform in Kenya’s political environment. The mean return on assets (7.4%) implies healthy profitability for most firms despite occasional political volatility.

To assess initial relationships between the core variables, Table 4.2 reports the Pearson correlation matrix. This table helps determine whether key relationships among political-regulatory stability, investment, and profitability are consistent with theoretical expectations.

Note.p < .05, p < .01.

As shown in Table 4.2, the Political and Regulatory Risk Index (PRRI) is positively correlated with capital expenditure (r = .47, p < .01), investment growth (r = .42, p < .01), and profitability (r = .36, p < .05). Conversely, PRRI is negatively correlated with cash holdings (r = −.41, p < .05), implying that firms operating under more stable political and regulatory conditions are more willing to invest and less inclined to retain excess liquidity. These descriptive patterns align with theoretical expectations that predict higher investment activity and profitability when institutional risk is lower.

4.2 Empirical Findings

This subsection presents the results of the econometric estimations, which formally test the relationship between political-regulatory stability (PRRI) and MNC performance outcomes. Two models are estimated: a Fixed Effects (FE) model controlling for time-invariant firm heterogeneity, and a Dynamic System Generalized Method of Moments (System GMM) model that incorporates lagged dependent variables to address potential endogeneity and persistence effects. The baseline fixed-effects results are shown in Table 3. This model assesses how variations in PRRI affect five dependent variables capital expenditure ratio, investment growth, cash ratio, leverage, and return on assets while controlling for firm size, firm age, ownership structure, and GDP growth.

Note. Robust standard errors in parentheses. p < .10, p < .05, p < .01.

The results in Table 4.3 show that PRRI is a significant predictor of corporate investment and performance outcomes. A one-unit improvement in political-regulatory stability increases the capital expenditure ratio by 6.4% and investment growth by 7.2%. Similarly, profitability (ROA) rises by approximately 3.1%. The negative coefficient for CashRatio (−0.043) indicates that firms reduce cash holdings when political stability increases, signaling higher confidence in the operating environment. Leverage remains statistically insignificant, suggesting that capital structure adjustments are less responsive to short-term political fluctuations.

To validate these results and account for potential dynamic behavior, a system GMM model was estimated. This approach controls for endogeneity by using lagged variables as instruments and incorporates the persistence of firm outcomes over time. Table 4 presents the results of this model.

Note. Robust standard errors in parentheses. p < .10, p < .05, p < .01.

As shown in Table 4.4, the lagged dependent variable (Y_t−1​) coefficients are positive and significant across all models, confirming persistence in investment and profitability behaviours among MNCs. The lagged PRRI (PRRI_t−1​) remains positive and statistically significant, reinforcing that past improvements in political-regulatory conditions have a continuing effect on firm outcomes. Diagnostic tests (Hansen J-test and AR(2)) indicate that the instruments are valid and there is no second-order autocorrelation, suggesting model robustness.

Discussion

This discussion places the empirical findings of this study showing that greater institutional stability (as measured by our Political and Regulatory Risk Index, PRRI) correlates with higher investment intensity, stronger growth, and lower precautionary liquidity among multinational corporations (MNCs) in Kenya into conversation with the broader academic literature on political risk, regulatory uncertainty and corporate behaviour. The goal is to interpret the results, relate them to theory, explore mechanisms, and reflect on implications for MNC strategy and Kenya’s governance environment.

Institutional quality, investment and risk-taking

The positive and statistically significant relationship between PRRI and capital expenditure ratio (CapEx_Ratio) found in our fixed-effects and dynamic panel models echoes established research showing that better governance and lower political risk reduce the hurdle for irreversible investment. For example, Political institutions and corporate risk‐taking: International evidence (Bao & Cardoza, 2023) find that sound political institutions across 90 countries are associated with greater corporate risk-taking, which is consistent with our finding of higher investment in years of stronger regulatory quality. The underlying logic is rooted in real-options theory: when institutional uncertainty falls, firms are more willing to commit rather than wait (Dixit & Pindyck, 1994). Our results therefore support the view that institutional stability constitutes a key location advantage under the OLI paradigm (Dunning, 1993; see also MANAS Social Research Journal, 2021).

Moreover, the finding that investment growth (InvGrowth) also responds positively to PRRI further suggests that not only are firms choosing to invest when risk falls, but that they are able to expand their asset base more dynamically. This parallels the literature on foreign direct investment (FDI) in Africa: for example, Political Risk, Volume of FDI, and Ownership Strategy: Contextual Analysis of African Markets (Zakari et al., 2022) document that while high-risk countries remain more difficult for FDI, where stability is higher MNCs invest more and prefer ownership structures with greater control. Our Kenyan-based evidence extends that literature to firm-level outcomes and shows how political/regulatory improvements unlock not just entry but growth and reinvestment.

Liquidity, leverage and profitability responses

The negative coefficient on the CashRatio (i.e., firms hold less cash when PRRI is higher) corresponds with theoretical expectations under uncertainty: when political or regulatory risk declines, firms reduce the “insurance” fraction of their balance sheet and allocate more assets toward productive uses. This aligns with the findings of Firm‐level political risk and implied cost of equity capital (Mishra, 2023) who document that firms with higher political risk face higher cost of equity and therefore reduce investment and increase precautionary liquidity. In our sample, higher PRRI (i.e., lower risk) means lower cash holdings: the firms feel less need to hoard liquidity, consistent with resource-dependence theory (Pfeffer & Salancik, 1978) which predicts that firms adjust their financial structure in response to exogenous dependencies and risk exposures.

Leverage exhibited a weaker response to PRRI in our results; although the coefficient was positive it was not statistically significant. This suggests that debt structure may be less sensitive to institutional shifts in the short- to medium-term perhaps because debt obligations are sometimes based on long-term contracts or host-country incentive regimes which themselves change slowly. Empirical work on leverage and political risk is still nascent, though some international studies (e.g., Gyimah et al., 2022) observe shifts in capital structure under governance stress. Thus, our finding of non-significant leverage change may reflect the Kenyan context: MNC affiliates perhaps had adequate access to finance even under moderate risk, or shift to equity rather than debt when risk changes.

Profitability (ROA) improved when PRRI was higher. This is consistent with theoretical expectations: better governance reduces transaction and agency costs, enhances contract enforcement and reduces firm‐level uncertainty, thereby improving operational efficiency (North, 1990; Farooq et al., 2022). Empirical work on political risk and profitability less abundant than investment studies also finds a positive link. For example, firms with less exposure to political uncertainty show higher return on assets and lower cost of capital (Hassan et al., 2019). In the Kenyan context, our findings suggest that improved regulatory quality and institutional stability not only support investment, but also convert into improved performance once investment is made.

Mechanisms: Why does institutional stability matter?

There are several plausible mechanisms through which higher PRRI (i.e., stronger institutional/regulatory environment) furthers investment and performance among MNCs in Kenya. First, regulatory predictability reduces the option value of waiting and encourages irreversible commitments (real options). When licensing, tax, environmental and procurement rules are stable, firms perceive lower risk of policy reversal, expropriation or regulatory renegotiation (Busse & Hefeker, 2007). Our descriptive evidence showing CapEx_Ratio declines around election years and spikes in years of reform supports this mechanism.

Second, improved governance enhances contract enforcement, rule of law and institutional capacity to manage disputes. Firms are more willing to invest when they believe legal frameworks will protect their rights and when corruption and informal demands are lower (Henisz, 2000). In Kenya’s case, the improvement in WGI Regulatory Quality and Rule of Law components during the middle of our period likely generated a better investment climate for MNCs. Third, improved institutional stability reduces uncertainty about cost of capital. Mishra (2023) show firm-level political risk increases cost of equity. In Kenya, when PRRI is higher, cost of capital may fall, enabling more investment. Our results notably the stronger investment and profitability responses suggest this channel is active.

Fourth, greater stability allows MNCs to shift from defensive to growth-oriented strategies. Our finding of reduced cash holdings under higher PRRI suggests firms shift from hedging (liquidity hoarding) to deployment (investment). This behavioural response aligns with institutional theory’s premise that firms adapt to the environment in order to gain legitimacy and reduce regulatory exposure (DiMaggio & Powell, 1983; Frynas & Mellahi, 2003). In Kenya’s case, MNC affiliates may feel more confident to expand rather than contract when rules are clearer and enforcement credible.

Kenya occupies an interesting niche: as an East African hub, it offers market size, regional connectivity and a relatively open investment regime but it also experiences electoral cycles, regulatory shocks and governance challenges. The empirical results from our study highlight that institutional improvements matter in this context: during periods of higher PRRI, MNCs increased investment, growth and profitability; during election years or regulatory upheaval, firms throttled investment, increased cash holdings and were less willing to grow. The descriptive patterns we observed investment declines in election years, increased cash ratios in high-risk years mirror prior country-specific studies of Africa (Mwega & Ngugi, 2006).

These results suggest that Kenyan policymakers seeking to attract and retain multinational investment should prioritize regulatory predictability, contract enforcement, streamlined licensing, and minimizing abrupt tax or procurement reforms. For MNC managers, the findings imply that political-institutional risk should be treated as more than a peripheral concern: it directly affects investment timing, financing, liquidity strategies and performance. The fact that we find consistent effects across both investment and profitability outcomes strengthens the case for integrating institutional risk into strategic planning.

Conclusion

This paper examined how political and regulatory stability influences the investment behaviour and performance of multinational corporations (MNCs) operating in Kenya between 2010 and 2024. Drawing on firm-level secondary data and macro-level governance indicators, the analysis demonstrates a clear and statistically significant relationship between institutional quality and corporate outcomes. Specifically, improvements in the Political and Regulatory Risk Index (PRRI) are associated with higher capital expenditure, greater investment growth, and enhanced profitability, alongside lower precautionary cash holdings.

These findings highlight the extent to which MNCs’ strategic and financial choices are conditioned by the predictability and credibility of the host country’s governance environment.The study’s results contribute to the broader political economy and business literature in several important ways. First, they empirically validate the long-standing theoretical claim that political and regulatory stability lowers uncertainty, thereby encouraging irreversible investment and expansion by foreign firms. Second, they show that institutional quality does not merely influence market entry decisions but also shapes ongoing operational behaviour particularly in how MNCs allocate resources and manage liquidity under varying degrees of risk. Third, the study provides evidence from Kenya, an emerging market that has become an anchor economy in East Africa but continues to experience periodic political and policy disruptions.From a policy perspective, the results underscore the critical role of governance reform in sustaining investment-led growth. Policymakers seeking to attract and retain multinational capital must focus on strengthening the rule of law, regulatory predictability, and contract enforcement. A transparent and stable regulatory environment reduces transaction costs and signals long-term commitment to investors. For corporate managers, the findings emphasize the strategic importance of political-risk assessment and proactive engagement with institutional processes. Firms that align their strategies with evolving governance realities are more likely to achieve resilience and profitability in emerging markets.

7. Limitation

While the results are robust, they should be interpreted in light of certain limitations. First, institutional indicators such as WGI capture country-level conditions and may mask firm-level or sector-level heterogeneity in risk exposure; as Hassan et al. (2019) show, firm-level political risk varies significantly even within the same country. Second, our study focuses on MNC affiliates in Kenya only; while this provides depth, extension to cross-country samples in sub-Saharan Africa would help generalise findings. Third, although we used dynamic panel and DiD designs, residual endogeneity (e.g., reverse causality from firm performance to regulatory environment) cannot be fully eliminated; future work might exploit more granular event-data or quasi-natural experiments for causal inference. Finally, future research could examine how ownership structure, local partnerships and CSR engagements moderate the relationship between institutional risk and firm behaviour (see Adeyeye, 2012; Osuji et al., 2019).

Appendix

Appendix A. Data Sources, Variables, and Transformations

A1. Data Compilation Framework

The analytical dataset was compiled from multiple secondary sources integrating firm-level, macroeconomic, and institutional data for Kenya covering the period 2010–2024.

·       Firm-level financial data were obtained primarily from audited annual reports filed with the Nairobi Securities Exchange (NSE, 2024) and from the Orbis Global Database (Bureau van Dijk, 2025).

·       Where listed-company data were incomplete, supplementary indicators for foreign-owned private firms were obtained from the UNCTAD Foreign Direct Investment (FDI) Statistics (UNCTAD, 2023), which track multinational affiliates by country and industry.

·       Macroeconomic variables such as GDP growth and inflation were drawn from the World Bank (2025) World Development Indicators (WDI), while governance measures were sourced from the World Governance Indicators (WGI, 2024).

·       Exchange-rate and CPI series were retrieved from the Central Bank of Kenya (CBK, 2025) and the Kenya National Bureau of Statistics (KNBS, 2025), respectively.

·       Political instability data including protests and violent events were obtained from the ACLED (2025) Kenya country file.

All files were downloaded between June and August 2025, verified for completeness, and harmonized into a single unbalanced panel of 950 firm-year observations.

A2. Variable Construction and Transformations

Recognizing that firm financial behaviour exhibits temporal persistence, a dynamic panel model was estimated to incorporate lagged dependent variables, expressed as:

Here,  measures the degree of adjustment inertia. Because the lagged term introduces potential endogeneity, estimation relied on the System Generalized Method of Moments (GMM) technique (Arellano & Bover, 1995; Blundell & Bond, 1998). The validity of instruments was confirmed using the Hansen -test, while the Arellano-Bond  statistic ensured the absence of second-order serial correlation.

For regulatory reforms affecting only certain industries, the study employed a difference-in-differences (DiD) estimator to isolate causal effects of specific regulatory changes. The model takes the form:

where Treated  equals 1 for firms in affected sectors and  equals 1 for years after the reform. The coefficient  captures the average treatment effect of the regulatory shock on the treated group. Pretreatment parallel trends were verified through graphical and statistical tests based on event-study estimations of leads and lags.

Endogeneity concerns were further addressed through an instrumental variable (IV) framework using exogenous regional political shocks as instruments. Specifically, the number of political unrest events in neighbouring countries (Uganda, Tanzania, and Ethiopia) served as instruments for domestic instability, under the assumption that cross-border unrest affects Kenya's perceived political risk but not individual firm performance directly. The first-stage regression tested instrument strength via the F-statistic, while the second-stage regression estimated the structural parameters.

Appendix B: Sample Multinational firms

Appendix C. Regulatory Reform and Event Coding

C1. Identification of Regulatory Events (2010–2024)

C2. Coding Framework

Two dummy variables were constructed to capture institutional and policy shocks relevant to multinational firms:

1.     ElectionYear – equals 1 in years when Kenya held national general elections (2013, 2017, 2022) and 0 otherwise, based on official data from the Independent Electoral and Boundaries Commission (IEBC, 2025).

RegulatoryShock – equals 1 for fiscal years in which major sector-specific policy or licensing changes occurred, identified through government and regulatory records (Kenya Gazette Notices, CBK Circulars, and EPRA Directives)

C3. Coding Method

·       Each firm-year observation was matched to the relevant sectoral reform year.

·       RegulatoryShock = 1 if the firm operated in a sector affected by any of the above events during that year.

·       Otherwise, RegulatoryShock = 0.

·       Election years were coded separately to allow difference-in-differences analysis of political vs regulatory shocks.

C4. Validation and Cross-Checking

·       Reforms were validated through cross-reference with Kenya Law Reports, CBK policy archives, and EPRA Annual Reports (2012–2024).

·       Sector classifications followed the International Standard Industrial Classification (ISIC Rev. 4) codes used in the Orbis dataset.

·       Inter-coder reliability was checked by two independent research assistants; Cohen’s κ = 0.91 indicated high consistency in event coding.