A puzzle hiding in plain sight
Pick up the annual report of a large American company and flip through it. You will likely find hundreds of pages covering business segments spread across continents, derivative contracts, pending lawsuits, restructuring plans, and layers of subsidiaries. Some reports read like encyclopedias. Others are relatively slim.
Does all that corporate complexity change how much a company borrows? A new investigation published in the Journal of Business Research sets out to answer that question. The short version of the finding: companies whose disclosures reveal more operational and informational complexity tend to carry more debt, and the pattern is strongest among firms that usually have the hardest time raising money.
The question behind the research
Corporate finance textbooks have long explored why companies pick their mix of debt and equity. Classic theories point to tax benefits, the risk of bankruptcy, and the information gap between managers and outside investors. Yet one trait of modern firms has stayed mostly in the background: their sheer complexity.
Rudresh Pandey, an associate professor at Oslo Business School at Oslo Metropolitan University, wanted to bring that trait into the foreground. He notes that complexity is a “double-edged sword” for lenders. On one side, a company operating in many regions and product lines may have steadier overall cash flows, making it a safer bet. On the other, such a company can be harder to analyze, raising the cost of monitoring and the chance of nasty surprises.
Because these forces push in opposite directions, theory alone cannot predict which one wins out. That is what makes the question an empirical one.
Defining what “complexity” actually means
Pandey is careful to distinguish information complexity from related ideas. Diversification is about how broadly a company operates. Opacity is about deliberate secrecy. Disclosure quality is about clarity. Information complexity, as used here, is about the volume and interconnectedness of the information a company produces, even when it is being fully transparent. A company can be perfectly honest and still hard to understand.
To measure it, he relies on a text-based index developed by researchers Tim Loughran and Bill McDonald, who scanned 10-K filings (the detailed annual reports U.S. public companies file with the Securities and Exchange Commission) for words that tend to appear in complex business descriptions. Terms like “derivative,” “litigation,” “restructuring,” and “worldwide” feed into a complexity score. To guard against confusing complexity with simple verbosity, Pandey also includes the logged file size of each filing as a separate measure.
How the study was built
The sample covers U.S. public firms from 1996 through 2021, excluding financial companies and utilities because they operate under unusual capital rules. That leaves about 35,548 firm-year observations drawn from Compustat, a standard financial database.
Pandey ran regressions linking complexity to several measures of leverage: total debt divided by book value of assets, total debt divided by market value, short-term debt, long-term debt, and total liabilities. He controlled for profitability, tangible assets like property and equipment, market-to-book ratio, firm size, whether the company was audited by a Big 4 firm, and whether it reported foreign income. He also included year and industry fixed effects, which strip out the influence of economy-wide shocks and industry-specific patterns.
What the numbers showed
Across every measure of borrowing, complexity came out positively and significantly linked to higher leverage. More complex firms, in other words, carried more debt. Disclosure volume (the logged file size) pointed the same way. The traditional controls behaved as expected: more profitable firms borrowed less, while firms with more tangible assets borrowed more.
To push beyond correlation, Pandey used three additional strategies. First, he took advantage of the SEC’s staggered rollout of its XBRL filing mandate between 2009 and 2011, which required larger firms to adopt a more structured digital reporting format earlier than smaller ones. This created an outside shock to disclosure complexity. Comparing treated and untreated firms before and after, he found that the mandated jump in complexity was followed by modest but statistically meaningful increases in leverage.
Second, he ran models using complexity measured one and two years before the leverage figures, which helps address the concern that borrowing somehow causes complexity rather than the other way around. The link held.
Third, he used a matching technique called propensity-score matching to pair high-complexity firms with otherwise similar low-complexity firms. A year later, the high-complexity group had book leverage 3.5 percentage points higher and market leverage 2.4 percentage points higher than their matched counterparts.
A further test suggested the relationship is not purely linear. Adding a squared complexity term produced an inverted-U shape: leverage rises with complexity up to a point, then falls. Very high complexity appears to tip lenders toward caution.
The financial constraints twist
Pandey then split the sample into financially constrained firms (smaller, without dividends, or with weaker operating cash flows) and unconstrained ones. The complexity-leverage relationship was roughly twice as strong for constrained firms. For the size split, the complexity coefficient was 0.152 for constrained firms compared with 0.064 for unconstrained ones.
One reading is that complexity works hardest as a borrowing lever precisely when borrowing is hardest to come by.
What this might mean for businesses
For managers, the study suggests that how a company structures and communicates its operational breadth can influence its access to debt. Complexity paired with credible disclosure appears to build creditor confidence. For lenders and analysts, quantitative complexity measures drawn from filings may add useful texture to credit assessments.
Pandey adds important caveats. Complexity that outstrips a firm’s ability to manage or explain itself can weaken internal controls and raise the risk of over-leverage. And the inverted-U pattern is a reminder that more complexity is not automatically better. The study describes an association observed in a large U.S. sample; whether the same dynamic holds in other institutional settings remains an open question.
