Economic Drivers of Public Health Expenditure in Kenya

Studies Management and Finance Economics, of Journal 0504-2644 (online): ISSN 0490,-2644 (print): ISSN 5202 October 10 Issue 80 Volume 8.317 Factor: Impact ,15-i10-10.47191/jefms/v8 DOI: Article 7226 -6711 No: Page JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6711 Economic Drivers of Public Health Expenditure in Kenya Dr. Matundura Erickson1, Dr. Naftaly Mose2, Sam Abeid3, Nyaboga Cynthia Kwamboka4 1,2,3,4Department of Economics, University of Eldoret,P.O Box 1125-30100, Eldoret Kenya ABSTRACT: Health spending is a major concern in low and middle income countries due to less financing to the health sector. One of the main goals of the Kenyan government's "big four" development strategy, which is scheduled for completion by 2022 and was achieved in some few counties in 2023 is universal health care. Health has consistently been prioritized over time and has occupied a central position in political campaign platforms. The government has consistently spent huge amount of money into the health sector. In Kenya, majority of people depend on public insurance and only a very small portion of Kenyans can afford to have access to the private insurance and out of pocket payment, this has led to increased level of poverty and higher dependency ratio. Despite these efforts, Kenya continues to face challenges in effectively allocating public health expenditure with macroeconomic factors playing a significant role in influencing spending patterns. However, existing studies have not adequately examined how corruption, unemployment, and fiscal deficit impact public health expenditure in the country. This study sought to fill this gap by providing empirical evidence on these relationships. The purpose of this study was to ascertain how macroeconomic factors affected Kenya's public health spending. This study aimed to establish effects of GDP per capita, corruption, unemployment fiscal deficit and tax revenue on in Kenya. The key theoretical anchors of the study are Public Expenditure theory and Wagner’s theory. Explanatory research design was used. Secondary data from the Kenya National Bureau of statistics (KNBS) was used with annual time series data spanning from 1990 to 2023. The data was subjected to stationarity test using Augmented Dickey Fuller (ADF) test, Phillips and Perron (PP) and Kwiatkowski–Phillips–Schmidt–Shin (KPSS) for unit root test. The study employed Autoregressive Distributed Lag model (ARDL) to evaluate the relationship among the variables. The long run ARDL analysis revealed that the coefficients public health expenditure of; corruption -2.231 (p-value 0.002<0.05), per capita gross domestic product 0.001(p-value 0.02<0.05), tax revenue 0.075(p-value 0.025<0.05), and unemployment 0.227(p-value 0.03<0.05) significantly affected public health expenditure in Kenya. However, the fiscal deficit was found to be insignificant in the long run 0.008(p-value 0.914>0.05).To ensure prudent public health expenditure in Kenya, the study recommends strengthening anticorruption laws, maintaining fiscal discipline through effective budgeting, promoting per-capita economic growth by boosting productivity and investments. Optimizing tax revenue through efficient policies and broadening the tax base is vital to fund public services. Addressing unemployment by creating jobs and investing in education is crucial for effective use of the labor force. KEYWORDS: public health expenditure, GDP per capita, Corruption, Unemployment, Fiscal deficit, Tax revenue, ARDL INTRODUCTION Global Public Health Expenditure Overview According to the World Health Organization (WHO) report, there is still a significant disparity in health spending across the globe, with over 80% of people living in lowand middle-income countries while only making up 20% of global health spending. Lowand middle-income countries (LMICs) have significant obstacles in attaining Universal Health Coverage due to allocation of less health financing. This results in high out-of-pocket health expenditure and poor health services (Behera & Dash, 2019) WHO global expenditure report (2018) on health, shows that in low-and middle-income countries, out-of-pocket payments have remained high, accounting for more than 40% of total health spending in 2018. The rapid growth of public health expenditure has become a great concern for both household and governments. Both governments and households are growing more concerned about the rapid increase in health costs. According to Jakovljevic et al. (2020), global health spending has increased over the last two decades by double in real terms, reaching US$ 8.5 trillion in 2019 and 9.8% of GDP which is up from 8.5% in 2000. The distribution of health spending is still more Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6712 uneven than that of the global GDP. The United States alone represented over 40% of global health expenditure, with high-income nations making up around 80% of this amount. On average, their per capita spending on health was more than four times higher than the GDP per capita of low-income countries. About half of the health spending in lowand middle-income countries went toward primary healthcare, accounting for roughly 3% of GDP on average. About one-third of primary healthcare spending came from government sources, and the other half came from external aid. Africa Public Health Expenditure Overview According to World Health Organization (2010), Africa is increasing its investments in healthcare to improve health outcomes and progress toward achieving the Sustainable Development Goals (SDGs). Initiatives like the 2001 Abuja Declaration highlight the commitment of African leaders to prioritize health in national development. However, despite these efforts, the continent faces challenges such as inadequate funding, with health spending in African countries far behind that of high-income nations. In 2010, Africa's average health spending was just US$ 135 per capita, significantly lower than the US$ 3,150 in high-income nations. The global economic crisis has added further strain on health spending, especially in lowand middle-income countries, making it difficult to meet health expenditure targets. The main obstacle to improving healthcare in Africa is the insufficient funding and ineffective health financing systems. In many African nations, household out-of-pocket expenses account for a large portion of healthcare costs, often leading to financial hardship for individuals. While some countries, like Malawi, have received substantial foreign assistance, overall, less than 20% of health funding in most African countries comes from external sources. The continent's healthcare systems are underdeveloped, and a significant budget gap for public health, estimated at $66 billion annually, remains. To address these challenges, African governments must increase domestic healthcare funding and reduce reliance on out-of-pocket payments. Efforts like the Abuja and Maputo Declarations have set targets for health sector funding, but many countries are still struggling to meet them Mouteyica and Ngepah (2023) Public Health Expenditure in Kenya Kenya's public health expenditure plays a crucial role in healthcare financing accounting for 46% of total health spending, while private sector and donor contributions make up the rest. Despite the existence of public healthcare coverage through the National Hospital Insurance Fund (NHIF), inefficiencies such as weak administration, low claim settlement rates, and limited accessibility for the economically disadvantaged hinder its effectiveness. Out-of-pocket (OOP) expenditures remain a major concern, making up a significant portion of private health spending, which has increased over time (Barasa et al., 2017).Health financing challenges have led to fluctuating user fee policies in public hospitals, impacting access to healthcare, particularly for low-income populations. Kenya's healthcare system remains underfunded, with public health spending falling well below the Abuja Declaration’s recommended 15% of GDP, standing at just 1.5% in 2012 (Nyamwange (2012). The government has recognized healthcare as a fundamental right, prioritizing universal health coverage (UHC) in national development plans, including Vision 2030 and the Big Four Agenda (Moon et al., 2016). However, macroeconomic factors such as GDP per capita, fiscal deficits, unemployment, and corruption continue to influence healthcare funding and accessibility (Organization, 2013). Donor funding for health has declined, while reliance on OOP payments has increased, exacerbating financial hardships for citizens. Addressing these challenges requires a more sustainable health financing model, increased government investment, and better resource allocation to ensure equitable access to quality healthcare services for all Kenyans (SeitioKgokgwe et al.) 1.2 Statement of the Problem Kenya's high reliance on out-of-pocket (OOP) healthcare payments remains a significant challenge with 26.1% of total health financing coming from OOP expenses in 2017, compared to much lower rates in countries like Seychelles, Botswana, and South Africa (World Bank, 2018). Such spending forces many households into poverty by reducing their ability to afford other essential goods and services (Nkatha, 2019). Despite efforts to improve public health expenditure, factors such as corruption, fiscal policies, unemployment, GDP per capita, and tax revenue continue to limit resource mobilization and hinder the development of the health sector (Osoro, 2015). While healthcare is a fundamental right, many lowand middle-income countries, including Kenya, allocate insufficient government funding to public health, leading to inadequate service provision and increased financial burdens on citizens (Bein, 2020). Achieving Sustainable Development Goal (SDG) 3, ensuring healthy lives and well-being for all requires addressing Kenya’s persistent healthcare financing gaps. Despite progress in improving health outcomes, limited government funding and high OOP payments continue to restrict access to quality healthcare services. The failure of the National Hospital Insurance Fund (NHIF) to Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6713 provide comprehensive coverage further exacerbates the issue, leaving many uninsured and vulnerable (World Bank, 2018; (Nkatha et al., 2020). To address these challenges, this study seeks to examine the impact of macroeconomic drivers; fiscal deficit, unemployment, GDP per capita, corruption and tax revenue on public health expenditure in Kenya, providing insights into strategies for sustainable health financing. 1.3.2 Specific Objectives of the Study The specific objectives of the study were; i. To evaluate the effect of GDP per capita on public health expenditure in Kenya ii. To determine the effect of tax revenue on public health expenditure in Kenya iii. To establish the effect of fiscal deficit on public health expenditure in Kenya iv. To assess the effect of corruption on public health expenditure in Kenya v. To determine the effect of unemployment on public health expenditure in Kenya 1.5 Significance of the Study This study will benefit policymakers, insurance providers, and other health industry stakeholders by guiding the development of strategies to increase the uptake of insurance services in Kenya. It will also help insurance companies create plans to boost enrollment, particularly among individuals in the formal sector. The findings will support the Kenyan government in formulating policies that promote medical insurance uptake. Further, the study will contribute to health economics literature, providing insights that can enhance existing research and inform future studies. LITERATURE REVIEW 2.1.1 Public Expenditure Theory Public expenditure refers to government spending on essential services, including public health, infrastructure, and social welfare. Lewis (1952) emphasized the need for optimal fund allocation to ensure maximum returns while Keynes later highlighted the role of macroeconomic factors in influencing government expenditure trends. Over time, public spending has increased as governments recognize its importance in addressing market failures and supporting economic development. However, Brennan (2008) argues that correcting market failures is complex and cannot always be achieved solely through direct public provisions Brennan (2008). Brennan (2008) noted that despite its significance, the theory of public expenditure has limitations. The external impacts on consumption challenge its core premise, the world policy decisions involve more than simple binary choices and economists lack the authority to assign definitive social welfare weights. This underscores the need for a structured approach to public health spending, ensuring that resource allocation balances economic efficiency and social well-being. Fiscal Deficit and Public Health Expenditure Behera and Dash (2019) found that in the short term, Kenya's public health expenditure (PHE) benefits from economic growth and domestic borrowing, as borrowing helps expand health sector investments. However, other studies suggest that increased debt can reduce current public health spending over time. Unfavorable macro-fiscal policies have limited resource mobilization, slowing PHE growth and hindering health sector development. In as much as wealthier nations prioritize health spending under stable fiscal policies, economic crises often lead to reduced allocations. Some countries, like those in the former Soviet Union, have mitigated such impacts by securing external grants and optimizing health budgets, preventing budget cuts even during financial crises. Kenya’s health sector has struggled due to unfavorable macro-fiscal policies that limit economic resource mobilization and slow overall sector development. The fiscal deficit has strained health spending, as the gap between total revenue and expenditure forces the government to rely on borrowing. While borrowing may provide short-term relief, Behera and Dash (2019) caution that long-term debt obligations like interest payments reduce available funds for essential services like healthcare threfore weakening PHE sustainability. 2.2.2 GDP per capita and Public Health Expenditure Rono (2013) found that GDP significantly influences public health spending in Kenya, with a unit change in GDP increasing health expenditure by 0.011 units, while external financing decreases it by 0.304 units. GDP per capita plays a crucial role in determining healthcare spending growth, as seen in OECD countries, where income elasticity is consistently above one. Behera and Dash (2019) also confirmed a positive correlation between GDP per capita and public health expenditure (PHE), noting that a 1% increase in income per capita leads to a 0.020% rise in PHE. However, economic slowdowns often result in reduced budget allocations for Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6714 healthcare due to fiscal constraints. (Nyamwange, 2012) further supports this view, showing that GDP per capita accounts for over 92% of changes in primary healthcare expenditure (PHCE) in developing countries, with income being the strongest determinant of health sector allocations. Okunade (2005) emphasizes that healthcare spending is highly income-elastic, meaning that as incomes rise, health expenditures grow proportionally or even more. Using data from 30 African countries, Okunade found that income elasticity of health spending is close to unity, highlighting disparities in health expenditures across nations due to variations in economic structures, demographics, and governance. These disparities contribute to differences in health outcomes among countries. Overall, the studies suggest that economic growth positively influences public health spending, but external financial dependence and fiscal constraints can hinder sustainable healthcare investment. Unemployment and Public Health Expenditure Unemployment in Kenya significantly affects health expenditure, as those without jobs often lack access to private insurance, which covers most out-of-pocket healthcare costs. The country's small formal employment sector, low savings, and underdeveloped financial infrastructure limit options for both public and private health (Manda et al., 2020).As a result, many unemployed individuals are unable to afford medical care, increasing reliance on inadequate public healthcare services. Employment status plays a crucial role in acquiring health insurance. Fronstin (2007) found that unemployed individuals were less likely to have medical coverage, with a household survey in Kenya revealing that 41% of retail and wholesale workers lacked health insurance. Similarly, Baek and DeVaney (2005) determined that households with a formally employed head were more likely to have insurance. Limited insurance coverage among unemployed and informal sector workers is primarily due to their lower disposable income, making healthcare access a challenge for a significant portion of Kenya’s population. Corruption and Public Health Expenditure Corruption significantly affects public health expenditure in Kenya, reducing the efficiency and effectiveness of healthcare delivery. Kagotho et al. (2016) found that Kenya ranked 139 out of 168 countries in the global Corruption Perceptions Index (CPI), with 43% of respondents perceiving corruption in the healthcare sector. Bribery is widespread, with 35% of Kenyans reporting having bribed healthcare providers. Such corruption leads to the misallocation of resources, increased healthcare costs, and poor health outcomes, particularly for low-income citizens who cannot afford alternative healthcare options. Corruption occurs when public officials misuse their power for personal gain, often reducing transparency in governance and worsening healthcare quality. (Munywoki et al., 2023) argue that corruption causes inefficiencies by increasing healthcare costs, lowering care standards, and widening disparities in access to services. When funds intended for healthcare are misappropriated, the system loses critical resources, making healthcare less accessible and unaffordable for many. Ultimately, corruption remains a significant barrier to achieving equitable health expenditure and improving public health outcomes in Kenya. Tax revenue and Public Health Expenditure Behera and Dash (2019)found a significant positive correlation between tax revenue and Public Health Expenditure (PHE), with a 1% increase in tax income leading to a 0.057% rise in PHE. Direct taxes had a positive impact, while indirect taxes negatively affected PHE, with a 1% increase in direct tax increasing PHE by 0.025% and a 1% rise in indirect tax decreasing PHE by 0.061%. Manda et al. (2020) emphasized that adequate healthcare funding improves service quality and health outcomes, but limited tax bases and inefficient collection restrict government healthcare budgets. Similarly, Chipunza and Nhamo (2023)found that Zimbabwe's fiscal capacity and GDP growth positively influenced PHE, but economic crises significantly reduced tax revenue and healthcare funding. In Kenya, tax revenue is the primary source of government funding for healthcare, yet allocations remain insufficient. Over 70% of the Ministry of Health’s recurrent budget is spent on salaries, leaving minimal funds for supplies and services (Seitio-Kgokgwe et al., 2016) Ndajiwo (2020) noted that tax-to-GDP ratios in African nations have increased but are hindered by weak tax compliance, illicit financial flows, and inefficient incentives, limiting public health spending. Despite these challenges, tax revenue remains crucial for sustaining and expanding healthcare services in developing nations. Knowledge Gap The government has reduced health sector funding, straining public hospitals. Over 80% of Kenyans depend on public healthcare funding, while only 20% can afford private medical care and few have private insurance. This study examines the impact of macroeconomic factors on public health expenditure (PHE) in Kenya, addressing a gap in existing research which, primarily focuses on developed countries. Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6715 Conceptual Framework Figure 2. 1 Conceptual Framework Source: Researcher 2025 RESEARCH METHODOLOGY Study Area The study determined the effect of macroeconomic drivers on public health expenditure for the period 1990 – 2023. Research Design According to Schoonenboom and Johnson (2017) a research design refers to the arrangement of methods established to gather and analyze data in a way that is relevant to the research question. This study used an explanatory design to determine macroeconomic drivers and public health expenditure. Data Source This study used secondary time series annual data, from the period 1990 to 2023 on public health expenditure, fiscal deficit, GDP per capita, unemployment, corruption and tax revenue. Data was generated from Kenya National Bureau of Statistics and World Bank. 3.6 Model Specification The study adopted a multivariate model that include fiscal deficit, GDP per capita, unemployment, corruption and tax revenue written as 𝙿𝙷𝙴 = f (𝓕𝓓, GDP, U𝓝𝓜, 𝓒𝓡𝓣, 𝓣𝓡) Where; PHEPublic Health Expenditure, FDFiscal Deficit, GDPGDP Per Capita, UNMUnemployment, CRTCorruption, TRTax Revenue The study employed Autoregressive Distributed Lag (ARDL) Model because ARDL model accommodates different orders of cointegration hence it is consistent and efficient. Above Equation was modeled in ARDL as; ∇Yt = α + ∑βi∆𝑌 𝑡−𝑖 𝑝 𝑖=1 + ∑Ɣ𝑗∆𝑋𝑡−𝑗 𝑞 𝑗=0 + ɸ𝑌 𝑡−1 + ʎ𝑋𝑡−1 + εt……………….….…………..3.5 Where; ∇Yt= The change in the dependent variable 𝑌 at time 𝑡. 𝛼 = alpha, the constant term (intercept). βi =The coefficient of the lagged difference of the dependent variable ∆𝑌 𝑡−1. ∆𝑌 𝑡−𝑖= The change in the lagged dependent variable. 𝑝 = The maximum lag length of the dependent variable. Δ = Denotes the first difference, capturing short-run changes. Φ = the coefficient that represents the speed of adjustment towards the long-run equilibrium. It indicates how quickly the dependent variable returns to its long-run equilibrium after a change. ʎ = The coefficient of the lagged level of the independent variable 𝑋𝑡−1 Independent Variables Dependent Variable Fiscal deficit Unemployment GDP per capita Tax revenue Corruption Public Health Expenditure Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6716 Operationalization of Variables Table 3. 1: Description and Measurement of Variables Abbreviation Name of the Variable Description and Measurement Expected Sign Source PHE Public Health Expenditure The provision of health services is known as health expenditure, and it encompasses all costs related to nutrition, family planning, and health care. It will be measured by total expenditure on health as a percentage of gross domestic product (GDP) KNBS FD Fiscal Deficit Fiscal Deficit is the sum of domestic debt and external debt. It is measured as a percentage of the Gross Domestic Product Negative KNBS GDPP Gross Domestic Product per capita It is a measure of GDP divided by midyear population of the country. It is measured by taking real GDP divided by population Positive KNBS UNM Unemployment Refers when people who want to do work and are actively seeking jobs but cannot find employment measured by the number of unemployed as a percentage of the labor force. Negative KNBS CRT Corruption Refers to engaging in dishonesty or committing a crime with the intent of obtaining illegal benefits or abusing the position for their personal gain. It is measured by the Corruption Perceptions Index (CPI) Negative KNBS TR Tax Revenue Income that is collected by governments through taxation which is a source of government revenue. It is measured as a percentage of GDP. Positive KNBS µ𝑡 is the stochastic error term Factors that affect health expenditure but not captured in the model Source: Author’s Conceptualization, 2025 Unit Root Test Augmented Dickey-Fuller procedure was employed Model Diagnostic Tests A diagnostic test is crucial in examining a suitable model for establishing the relationship among the independent and dependent variables and how well the model fits. The following diagnostic tests were performed;Normality , Portmanteau, Multicollinearity and Heteroscedasticity Test Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6717 RESULTS, ANALYSIS AND INTERPRETATION Descriptive Statistics Table 4. 1 Descriptive statistics Variable Observations Mean Std. dev Minimum Maximum PHE 34 14.05328 15.08679 3.627142 42.73556 GDPACA 34 2874.668 1326.507 1704.031 6323.534 TAXR 34 16.89563 6.017338 13.26486 49.9 FD 34 -.8498235 2.524945 -4.661 3.495 CRPT 34 -1.010077 .1315275 -1.165813 - .7358834 UNMP 34 3.311647 .9807833 2.6 5.69 Source: Author (2025) Whereby; PHE=Public Health Expenditure, FD=Fiscal Deficit, GDPPACA= Gross Domestic Product Per Capita UNM= Unemployment, CRT=Corruption, TR=Tax Revenue Public health expenditure over a period of time has a mean of 14.05 percent and a standard deviation of 15.09 percent. Its minimum value is 3.63 percent and the maximum of 42.74 percent. Over the years, the rise in public health expenditure has effect on the economy's ability to mobilize resources, thereby hindering the growth of the health sector. The less financial resources allocated towards the health sector would lead to high out of pocket spending and hence poverty. GDP Per Capita parity had a mean of 2874.668 current US dollars with a standard deviation of 1326.507 US dollar. Its minimum has been 1704.031 current US dollars and a maximum of 6323.534 US dollars. Tax Revenue (TAXR) being one of the significant sources of funding for health has discovered a standard deviation of 6.02 percent with a minimum of 13.26 percent and a maximum of 49.9 percent. Tax revenue has a mean of 16.90 percent. Unemployment (UNMP) recorded a mean of 3.31 percent, minimum of 2.6 percent and a maximum of 5.69 percent. Fiscal deficit (FD) and corruption (CRPT) have the mean of negative values such as -.850 percent and -1.01 percent and a minimum of -4.67 percent and -1.17 percent respectively. While borrowing could seem like a brilliant idea in the short term, it might not be so good for Public Health Expenditure in the long run. This is due to the knowledge that paying down debt interest lowers the amount of funds available for the current government spending, such as healthcare as the mean showed a negative (-.850) percent (Behera & Dash, 2019) 4.2.1 Augmented Dickey-Fuller (ADF) Root Test Table 4. 2: Augmented Dickey Fuller Test for Unit Root at Levels and at First Difference Unit Root Test at Level Critical Values Variables Mackinnon pvalues Test Statistic 1% 5% 10% Remark PHE 0.3310 -1.893757 -3.646342 -2.954021 -2.615817 Unit root GDPACA 0.9922 0.782055 -3.646342 -2.954021 -2.615817 Unit root TR 0.0927 -2.658485 -3.661661 -2.960411 -2.619160 Unit root UNM 0.5634 -1.413426 -3.653730 -2.957110 -2.617434 Unit root CRPT 0.6752 -1.170493 -3.646342 -2.954021 -2.615817 Unit root FD 0.2380 -2.121405 -3.653730 -2.957110 -2.617434 Unit root Unit Root at First Difference PHE 0.0005 -4.816504 -3.653730 -2.957110 -2.617434 I (1) GDPACA 0.0153 -3.479141 -3.653730 -2.957110 -2.617434 I (1) TR 0.0000 -6.372384 -3.661661 -2.960411 -2.619160 I (1) UNM 0.0003 -5.077843 -3.661661 -2.960411 -2.619160 I (1) CRPT 0.0000 -6.035626 -3.661661 -2.960411 -2.619160 I (1) Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6718 FD 0.0026 -4.181539 -3.653730 -2.957110 -2.617434 I (1) Source: Authors Compilation from STATA Output, 2024 Upon first difference all Mackinnon p-values for the variables in the study were found to be below 0.0500, indicating stationarity: PHE (0.0005 < 0.0500), GDPACA (0.0153 < 0.0500), UNM (0.0003 < 0.0500), TR (0.0000 < 0.0500), CRPT (0.0000 < 0.0500), and FD (0.0026 < 0.0500). As a result, we fail to reject the alternative hypothesis and the null hypothesis of a unit root was rejected, demonstrating that the variables are stationary and integrated of order one (1). Determination Optimum Lag Length Table 4.7 identifies four lags as the optimal lag length, as it minimizes the selection criteria value. A lag that is too short may lead to autocorrelation in the error terms, which can distort statistical significance, making estimators appear relevant when they are actually inefficient. Proper lag selection ensures reliable and meaningful estimation results (Tiriongo, 2019). Table 4. 3: Optimum Lag Selection Criteria Lag LL LR Df P FPE AIC HQIC SBIC 0 -451.605 1.4e+06 31.2052 31.2948 31.4854 1 -292.879 338.4 36 0.000 210.724 22.3253 22.9528 24.2869 2 -240.498 104.76 36 0.000 96.9158 21.2332 22.3986 24.8763 3 -162.509 155.98 36 0.000 15.9622 18.4339 20.1373 23.7585 4 493.651 1312.3* 36 0.000 3.6e-16* -22.9101* -20.6688* -213.916* Source: Research Data, 2024 4.5 Bounds Test Based on this study’s results in Table 4.8, it is evident that a long-run relationship exists between the variables. The calculated Fstatistic of 9.179 exceeds the upper bound critical values at the 10%, 5%, 2.5%, and 1% significance levels, which are 3.35, 3.79, 4.18, and 4.68, respectively, at 𝐼(1). Hence, the null hypothesis of no level relationship was rejected. Table 4. 4: Bounds Test Critical Values (0.1-0.01), F-statistic, Case 3 F = 9.179, t = -4.763 [I_0] [I_1] L_1 L_1 [I_0] [I_1] L_05 L_05 [I_0] [I_1] L_025 L_025 [I_0] [I_1] L_01 L_01 k_5 2.26 3.35 2.62 3.79 2.96 4.18 3.41 4.68 accept if F < critical value for I (0) regressors reject if F > critical value for I (1) regressors k: # of non-deterministic regressors in long-run relationship 𝐻0: no levels relationship Critical Values (0.1-0.01), F-statistic, Case 3 [I_0] [I_1] L_1 L_1 [I_0] [I_1] L_05 L_05 [I_0] [I_1] L_025 L_025 [I_0] [I_1] L_01 L_01 k_5 -2.57 -3.86 -2.86 -4.19 -3.13 -4.46 -3.43 -4.79 accept if F < critical value for I (0) regressors reject if F > critical value for I (1) regressors Source: Authors’ Compilation from STATA Output, 2024 Diagnostic Checks Various diagnostic checks were conducted to ensure its reliability of the ARDL model used in the study Economic Drivers of Public Health Expenditure in Kenya JEFMS, Volume 08 Issue 10 October 2025 www.ijefm.co.in Page 6719 Multicollinearity Both the Breusch–Godfrey LM test and White's test for heteroscedasticity yielded p-values greater than 0.05, leading to the rejection of the null hypothesis. This indicates the absence of heteroscedasticity, autocorrelation, and multicollinearity in the model. Table 4.5: Auto Regressive Distributed Lag Model with Error Correction Term Sample 1994 – 2023 Log likelihood = 35.961934 Number of observations = 30 R-squared= 0.9436 Adj R-squared = 0.8366 Root MSE= 0.1264 D.exp Coef. Std. Err T P > |t| ADJ EXP L1 -.8834715 .1854703 -4.76 0.001 LR CORPN -2.230679 .5558094 -4.01 0.002 FD .00849 .0763173 0.11 0.914 GDPPACA .001268 .000458 2.77 0.020 TR .0754945 .0286606 2.63 0.025 UNM .2270987 .2182744 3.48 0.003 SR GDPPACA D1 -.0023892 .0008904 -2.68 0.023 LD .0001231 .000921 0.13 0.896 L2D -.0018797 .0006879 -2.73 0.021 TR D1 -.0531652 .013212 -4.02 0.002 LD -.0418354 .0109544 -3.82 0.003 L2D -.0240576 .0089636 -2.68 0.023 L3D -.0108146 .0054286 -1.99 0.074 UNM D1 -.0531652 .013212 -4.02 0.002 LD -.0418354 .0109544 -3.82 0.003 L2D -.0240576 .0089636 -2.68 0.023 L3D -.0108146 .0054286 -1.99 0.074 UNM D1 -1.144307 .270047 -4.24 0.002 LD -.4314789 .2580528 -1.67 0.125 L2D .1813071 .2716653 0.67 0.520 L3D -1.126417 .3012742 -3.74 0.004 TOT D1 -0.1115034 0.2285363 -0.49 0.632 LD -0.3114052 0.1853075 -1.68 0.111 FIR D1 0.0112124 0.0053738 2.09 0.052 LD 0.0089802 0.004639 1.94 0.070 CONS -4.821517 .895747 -5.38 0.000 Source: Authors’ Compilation from STATA Output, 2025 DISCUSSIONS AND FINDINGS The Effect of GDP per Capita on Public Health Expenditure in Kenya In the long run, GDP per capita has a positive and significant impact on public health expenditure, with a coefficient of 0.001268 (p-value 0.020). This means that a unit increase in GDP per capita leads to a 0.001268 unit rise in public health spending. The