Chilean SMEs gain more from personalized advisory support than from standardized training
A Chilean policy study draws a sharp line between two familiar forms of government support: personalized business advisory services and standardized training. Using administrative records linked to national financial data, investigators estimated how each approach affected investment, debt, and financial health in small and medium-sized enterprises. The comparison matters because both programs are designed to strengthen firms, but they do so through very different channels. Training spreads general knowledge across a broad audience; advisory services concentrate expertise on the specific constraints of one business at a time.
The analysis drew on roughly 100 baseline characteristics and financial histories, then used inverse probability of treatment weighting and post-double selection regression to reduce confounding in the observational data. In practical terms, that statistical design tried to mimic a more balanced comparison between firms that received one support modality and those that received the other. The result was not subtle: advisory services produced significantly higher investment value than training. In a setting where capital allocation is often the bottleneck between survival and expansion, the difference suggests that tailored guidance may unlock decisions that generic classroom instruction leaves untouched.
A single sentence captures the central divergence: personalized advice moved investment more than training did.
Strategic borrowing rose without a corresponding arrears signal
The financial pattern that followed advisory support is equally important. Firms associated with advisory services showed higher post-treatment debt, yet there was no significant increase in arrears. In a corporate finance setting, that combination often points to deliberate leverage rather than distress borrowing. Managers may have used improved information, tighter planning, or clearer growth strategies to take on debt as a tool for expansion, not as a desperate response to cash shortfalls. The debt increase therefore cannot be read in isolation; the absence of a parallel arrears signal makes the borrowing look more like calculated investment financing than a deteriorating balance sheet.
That distinction has direct policy implications. If support programs are judged only by debt accumulation, an intervention that helps firms borrow productively could be misclassified as risky. Yet the study’s financial outcome profile suggests that advisory support may improve the quality of borrowing decisions, not merely the quantity of borrowing. For agencies trying to decide where to place scarce public resources, the question is not whether debt rose, but whether debt was paired with healthier investment behavior and stable repayment performance.
Why firm-level tailoring matters more than one-size-fits-all support
The two support models likely work through different behavioral mechanisms. Standardized training can transfer general business principles, but it may stop short of changing the specific constraints that shape a firm’s financing decision. Personalized advisory services can interrogate cash flow patterns, investment timing, debt structure, and operational bottlenecks in a way that generic instruction cannot. For SMEs, especially those facing thin margins and limited internal financial expertise, customized guidance may lower decision costs and improve confidence around borrowing for growth. The observed investment response fits that mechanism: advice appears to have altered actual capital allocation rather than merely improving abstract knowledge.
The study also used post-double selection regression to examine heterogeneity in financial behavior, and female-owned SMEs stood out as more conservative. They invested less and borrowed less. Those patterns may reflect differences in risk tolerance, access to credit, prior exposure to external finance, or the distinct constraints female entrepreneurs face in capital markets. The paper does not reduce that finding to a single explanation, and it should not. Instead, it shows that firm ownership characteristics can shape the uptake and downstream effect of public support. A support program that works well for one segment of SMEs may underperform for another if the delivery model ignores those differences.
Female-owned SMEs displayed more restrained investment and borrowing behavior.
What the absence of an untreated control group does and does not allow
The study’s strongest claim is comparative, not absolute. Advisory services outperformed training when the two active support modalities were compared directly. Both programs may also have positive effects relative to no support, but the absence of an untreated control group prevents estimation of those absolute impacts. That limitation is central to how the findings should be read. The analysis can tell policymakers which of the two interventions appears more effective in this dataset; it cannot quantify how much either program improved outcomes over doing nothing.
Even with that constraint, the design is unusually informative for policy allocation. Government programs rarely operate in a vacuum, and program choice often comes down to whether agencies should expand broad training or concentrate resources in more individualized advisory services. This study points toward the latter when the objective is to raise investment value. It also suggests that debt creation alone should not be treated as failure if repayment behavior remains stable and borrowing is tied to investment activity. For ministries, development agencies, and local support programs, the practical question becomes whether their current menu of services is matching firm needs or simply distributing assistance evenly.
Chile’s administrative data make the policy signal harder to dismiss
The credibility of the findings rests in part on the data infrastructure. By integrating agency records with national financial data, the researchers could observe both program exposure and subsequent financial histories with far more depth than a small survey would allow. The baseline profile included around 100 characteristics, which gave the weighting and regression models a broad set of covariates to balance. That richness strengthens the argument that the observed differences are not just simple reflections of preexisting firm quality, even though no observational design can eliminate all residual confounding.
The takeaway for policymakers is narrow but actionable. Personalized advisory services generated a stronger investment response than standardized training, and they did so without a detectable rise in arrears. Female-owned SMEs showed more cautious financial behavior, signaling that program design may need to account for ownership structure as well as firm size and baseline finances. The remaining question is not whether government support matters; it is which firms need hands-on advice, which can benefit from standardized instruction, and how program managers should divide limited resources between those two paths.