The analysis relies on the WBES 2024 [3] dataset, which includes information from formal SMEs operating in trade, services, and light manufacturing. From the full dataset, a subsample of 27 firms was extracted based on the following criteria:

the firm is classified as a small or medium-sized enterprise (5 - 99 employees);the firm provided complete responses to innovation-related questions;the firm provided complete responses to institutional constraint indicators.

This subsample is consistent with the limited number of formal SMEs surveyed in the DRC and reflects the structural characteristics of the Congolese private sector.

Dependent variable: Frugal Innovation Frugal innovation is measured using the WBES item indicating whether the firm introduced a new or significantly improved product or service in the previous year, despite limited investment in formal R&D. This operationalization aligns with the literature on frugal innovation in resource-constrained environments.

Independent variables: Institutional Constraints Three institutional constraints were selected based on theoretical relevance and data availability:

Licenses and permits (severity of administrative procedures)Access to land (difficulty in obtaining or securing land)Regulatory delays (time required to comply with regulations)

Each variable is coded on a scale from 0 (“no obstacle”) to 4 (“very severe obstacle”).

Control variables Firm size, sector, and age were included as controls to account for structural differences among SMEs.

A firm is coded as a frugal innovator (1) if it answered “Yes” to WBES item D1 (“Has the firm introduced a new or significantly improved product or service in the last year?”) while reporting zero or very low R&D expenditure in item D7. Firms that did not introduce such innovations are coded as 0. This proxy captures the “do more with less” logic characteristic of frugal innovation.

The empirical analysis combines descriptive statistics and inferential tests:

Descriptive statistics to assess the prevalence of frugal innovation and the distribution of institutional constraints.Chi-square tests to examine associations between categorical variables.Independent samples t-tests to compare mean constraint levels between innovators and non-innovators.Binary logistic regression to estimate the probability of adopting frugal innovation as a function of institutional constraints and control variables.

Given the small sample size, statistical significance is interpreted cautiously, and emphasis is placed on effect sizes and directional trends rather than strict p-value thresholds.

The main limitation of this methodology is the reduced sample size (n = 27), which affects statistical power and generalizability. However, the WBES remains the most reliable dataset available for formal SMEs in the DRC, and the analysis provides valuable exploratory insights into innovation dynamics in highly constrained institutional environments.

The descriptive statistics presented in Table 1 summarize the institutional constraints faced by Congolese SMEs and the prevalence of frugal innovation within the sample. As shown in Figure 1Figure 1, approximately 70% of SMEs report adopting frugal innovation.

Table 1. Descriptive statistics of institutional constraints and frugal innovation.

Figure 1.Proportion of SMEs adopting frugal innovation.

Despite a complex and unfavorable institutional environment, the majority of Congolese SMEs surveyed implement forms of frugal innovation, whether in response to administrative delays, limited access to land, or heavy regulations. Although statistical analyses did not reveal significant links between each constraint and frugal innovation, the trend is consistent with the hypothesis: in a formally unfavorable context, Congolese businesses deploy adaptation, optimization, and ingenuity strategies typical of frugal innovation. This aligns with the findings of Leliveld, Knorringa, and Fort [13], who demonstrated, through case studies in the DRC and Kenya, that institutional constraints foster the local co-creation of frugal solutions, often based on informal knowledge and community-based approaches.

1) Logistic Regression

The coefficients are positive, which suggests that the more constraints an SME perceives, the more it tends to innovate frugally. However, no variable is significant (p > 0.05), so we cannot statistically conclude this with your sample.

Control variables (firm size, age, and sector) were included in the logistic regression but did not show meaningful effects. Their coefficients were small and statistically non-significant (p > 0.10), confirming that institutional constraints remain the main exploratory focus of the model.

As shown in Table 2, all coefficients are positive, suggesting that SMEs perceiving stronger constraints tend to adopt frugal innovation, even though none of the effects reach statistical significance.

Table 2. Logistic regression results on frugal innovation adoption.

Exploratory logistic regression: Odds ratios (OR) and 95% confidence intervals (CI) were computed to emphasize effect sizes: Licenses/permits constraint: OR = 1.065, 95% CI [0.94; 1.21]; Land-access constraint: OR = 1.009, 95% CI [0.85; 1.18]; Administrative time: OR = 1.018, 95% CI [0.92; 1.14]. All effects are positive but statistically non-significant (p > 0.05), consistent with the exploratory nature of the analysis.

2) Differences in Mean

Table 3. Differences in mean between frugal and non-frugal SMEs.

The statistical results presented in Table 3 suggest a positive association between the perception of institutional constraints and the tendency of SMEs to adopt frugal innovation strategies. The regression coefficients are all positive, indicating that the more constraints are perceived, the more firms tend to gravitate toward simple, robust, and accessible solutions.

However, statistical significance (p > 0.05) was not reached, which prevents any formal causal interpretation. This aligns with the findings of Chebbah [12], who argues that heavy regulations encourage SMEs to adopt non-standardized, often invisible but effective innovation processes. SMEs thus perceive these constraints as opportunities for innovation aimed at ensuring their survival in the market.

3) Overview of Statistical Findings

Table 4. Results of statistical analyses on frugal innovation.

Figure 2. Perceived licensing constraints among frugal and non-frugal SMEs.

The combined results presented in Table 4 summarize the logistic regression coefficients and the mean-comparison tests. All coefficients are positive, indicating that higher perceived constraints tend to be associated with a greater likelihood of adopting frugal innovation. However, none of these effects reach statistical significance (p > 0.05), confirming the exploratory nature of the analysis.

The boxplot in Figure 2Figure 2 compares the perception of licensing and permit constraints between frugal and non-frugal SMEs. A visual trend is observed, suggesting that firms adopting frugal innovation perceive licensing constraints slightly more acutely (higher median). However, this difference remains moderate and is not statistically significant (Student’s t-test, p = 0.229). This suggests that, despite the absence of formal statistical evidence, frugality may emerge as an adaptive strategy in response to a restrictive institutional environment.

Overall, all variables tested in the logistic model show positive effects on the propensity to innovate frugally, but none are significant at the 5% level. These results indicate a potential trend, but the limited sample size (n = 27) likely reduces statistical power. This supports the need for complementary qualitative approaches or future quantitative extensions.

Table 5. Chi-square test results for associations between frugal innovation and firm characteristics.

The results of the association tests presented in Table 5 show no statistical relationship between frugal innovation and firm characteristics such as size, sector, or location. The Chi-square test yields a pvalue of 1, indicating that the observed distribution is fully consistent with the assumption of independence.

The visual comparison further illustrates the absence of a clear pattern between perceived licensing constraints and the adoption of frugal innovation. Although a slight difference in medians can be observed, the variation remains moderate and statistically non-significant.

These results suggest that frugal innovation does not follow a rigid segmentation based on structural firm characteristics. Instead, the diversity of resilience contexts calls for the development of an exploratory typology grounded in qualitative or field-based approaches. This perspective highlights that frugality depends less on size or sector than on the degree of autonomy, creativity, and interaction with local constraints.

The results of the logistic regression model presented in Table 6 allow us to examine the relationship between the institutional constraints perceived by Congolese SMEs and their propensity to adopt a frugal innovation approach. All coefficients are positive, suggesting that the stronger the constraints, the more an SME tends to adopt frugal innovation.

Table 6. Logistic regression results on frugal innovation and institutional constraints.

Although these trends are consistent with the theoretical framework (Radjou et al., 2012; Agarwal & Brem, 2021), none of the variables reach statistical significance (p > 0.05). The constraint related to licenses and permits shows the highest coefficient (+0.063; p = 0.209), followed by administrative time (+0.0181; p = 0.677), while access to land exhibits only a marginal influence (+0.0095; p = 0.763). The model’s AIC (38.35) indicates a simple but informative structure, suitable for an exploratory study.

These results, while not formally validating the main hypothesis, support a conceptually robust interpretation. Frugal innovation can be understood as a driver of entrepreneurial resilience, deployed in response to a restrictive institutional environment. In the absence of stable formal resources, SMEs develop resourceful, creative, and adaptable solutions based on improvisation and local optimization.

Thus, although the identified constraints do not exert statistically significant effects, they appear to activate entrepreneurial adaptation strategies in which frugality becomes a tangible form of resilience. This dynamic justifies a more in-depth qualitative investigation of innovative behaviors in informal or sub-institutionalized settings, where emerging practices often defy traditional quantitative modeling.