A CRITIC-Weighted Fuzzy-TOPSIS and Machine Learning Framework for Evaluating Corporate Governance Efficiency: Evidence from Construction Materials Manufacturing in Uzbekistan
This study introduces a hybrid CRITIC-Fuzzy-TOPSIS and machine learning framework to evaluate corporate governance in Uzbekistan's construction materials sector, revealing a paradox where expert-perceived governance priorities differ from financial drivers, with relational factors like stakeholder relations proving more critical to profitability than structural dimensions.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of business, there is a long-held belief that how a company is run determines how well it performs. This idea, known as corporate governance, suggests that clear rules, honest reporting, and independent oversight create a stable environment where a company can grow and make money. In wealthy nations with strong laws and open markets, this connection is often taken for granted. However, in emerging economies where rules are still taking shape and information is harder to find, the relationship between good management and financial success is much murkier. Researchers have long wondered if the standard checklists used to judge a company's health actually predict its profits, or if they are merely measuring the wrong things. This question becomes especially urgent in places like Uzbekistan, where rapid industrial growth has outpaced the development of formal management systems, leaving many companies to operate with old habits and limited transparency.
A team of researchers from universities in Uzbekistan decided to test this relationship in a specific, vital sector: the manufacturing of construction materials like cement, bricks, and glass. They wanted to see if the way experts perceive a company's governance matches the reality of its financial performance. To do this, they built a new kind of evaluation system that combined human judgment with advanced computer learning. They gathered twenty experts who knew the industry well and asked them to rate eight major factories on ten different aspects of management, ranging from how well the board of directors functions to how the company treats its workers and interacts with the government. The experts used a simple seven-point scale, from "very poor" to "very good," to describe each company.
Once the experts provided their ratings, the researchers used a mathematical method to determine which of those ten aspects actually mattered most for distinguishing one company from another. They found that the experts were most able to tell the companies apart based on how well they planned for the future and how they managed risks. These two factors received the highest importance scores in the initial analysis. The team then used these scores to create a ranking of the eight companies, placing the one with the best perceived governance at the top and the one with the worst at the bottom.
Here is where the story takes a surprising turn. The researchers then compared this expert ranking against the actual financial records of the eight companies. They looked at a specific measure of profitability called Return on Assets, which calculates how much profit a company makes relative to the value of its fixed equipment and buildings. The results showed almost no connection between the expert rankings and the real money made. In fact, the relationship was slightly negative. The company that the experts rated as having the worst governance actually had the highest profits, while the company rated as having the best governance had one of the lowest profit rates. The computer models trained to predict profits based on the expert ratings failed to do so, confirming that the way experts see a company's management style does not tell you how much money that company is making.
To understand why this disconnect exists, the researchers used a technique that allows computers to explain their own reasoning. They asked the computer models to identify which specific factors, among the ten the experts rated, were actually driving the differences in profit. The answer was completely different from what the initial expert ranking suggested. While the experts focused on formal structures like strategic planning and risk management, the computer analysis revealed that the factors most closely tied to actual profits were the relationships a company had with people. Specifically, the quality of relations with employees and the nature of the company's dealings with government regulators were the strongest predictors of financial success.
This finding suggests that in the specific context of Uzbekistan's construction industry, the formal rules and documents that experts look for are not the primary drivers of daily success. Instead, the practical ability to keep workers motivated and to navigate the complex landscape of government regulations appears to be far more critical for the bottom line. The study does not claim that good governance is unimportant, but it does show that the standard way of measuring it—focusing on board independence and written plans—may be missing the most vital elements in this environment. The researchers conclude that for companies in this sector to improve their financial performance, they should prioritize building strong, constructive relationships with their workforce and the state, rather than simply polishing their formal management structures. This insight offers a new path for both business leaders and policymakers who are trying to understand what truly makes a company thrive in a developing economy.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.