Three regulators, three ways of handling other people's money

In Colombia, a company that handles other people’s money is supervised by one of three superintendencies, and which one applies depends neither on its size nor on its sector. It depends on whose money it handles.

Put the three universes in the same table — 32,817 entities and 3,752 trillion pesos — and a very clean staircase appears.

A note on scale: Colombian financial reporting uses the long scale, where billón means a million millions. Throughout this article “trillion” means 10¹², which is what a Colombian reader sees as billón.

The leverage staircase

From 46% to 87%. That is the one column that moves in a straight line with the level of supervision, and the reason is structural:

• A commercial company funds itself with its owners and its suppliers. The third-party money it handles is trade credit, and whoever extends it knows who they are extending it to.
• A cooperative entity also funds itself with its members’ savings. 63.4% of its liabilities are deposits from the very people who own it.
• A bank funds itself almost entirely with deposits from the public. For every 100 pesos of assets, 87 belong to people who have no way of assessing the institution’s risk.

The supervisory regime follows exactly that proportion. The more third-party money there is on the balance sheet, and the less its owner can assess it, the stricter the supervisor.

Which yields the most useful reading in the whole table: banking’s 87.3% leverage is not a sign of fragility. It is a description of the business. Reading it as “high risk” without that context is a reading error, not a finding.

Same ROE, ROA four and a half times apart

Commercial companies return 7.9% on equity. Banking, 7.4%. Almost identical.

On assets, the first returns 4.28% and the second 0.94%: 4.6 times less.

That gap is precisely the leverage. Banking reaches the same return for the shareholder by moving a far larger balance sheet with far less of its own money. That is the business: not charging more per transaction, but levering more.

It has a direct consequence for anyone assessing institutions: the ROE of a financial institution and that of a commercial company are not the same measure, even though they are calculated the same way. A 7% ROE backed by 13% equity and one backed by 54% equity are different risks carrying the same number.

The cooperative sector lands in the middle on leverage but last on both profitability measures. That is not an anomaly either: a cooperative does not exist to maximise its surplus, it exists to lend cheaply to its members. A low surplus is the objective met, not the problem.

A warning about the cut-off

The financial data covers half a year: January to June 2026. The other two are full years ended December 2025.

Annualised, banking’s ROE would sit near 14.8% and its ROA near 1.87%. The contrast between the two measures does not disappear, it widens: banking would return nearly twice as much to the shareholder as a commercial company, while moving a balance sheet where each peso of assets produces less than half as much.

Balance-sheet figures — assets, liabilities, equity, leverage — are comparable across the three, because they are snapshots at a date. Results for the period are not.

The least supervised regime is the one that loses most

22.6% of commercial companies closed 2025 in the red, against 16.1% of the financial sector and 12.1% of the cooperative sector.

Prudential supervision does not guarantee profitability — the cooperative sector is the least profitable of the three — but it does narrow the dispersion. Where there are minimum capital requirements, mandatory provisions and solvency margins, there are fewer entities at the bad end of the distribution.

254 entities weigh as much as 29,312

The 254 financial institutions hold 88% of the assets of the 29,312 commercial companies.

The average financial institution manages 6.82 trillion pesos. The average commercial company, 66,773 million. That is 102 times.

It is the strongest argument against adding the three universes into a single analysis: any aggregate that merges them ends up describing banking alone, because banking is 99% of the weight.

Why there are three search tools and not one

The three regimes are mutually exclusive by law. An institution supervised by the Superintendencia Financiera does not appear in Supersociedades, and across the three files there is not a single repeated tax ID. The sums in this table do not count anyone twice.

The charts of accounts, however, are different, and that is where comparability ends:

• Banking reports to the CUIF, with loan portfolios, deposits and derivative positions.
• The cooperative sector reports to the solidarity chart of accounts, with member contributions, mutual funds and permanent savings — line items that exist nowhere else.
• Commercial companies report under IFRS, full or for SMEs.

Assets, liabilities and equity mean the same thing in all three. The detail lines do not. That is why WiseData gives each regime its own search tool, with its own financial statement, instead of a single tool that would have to invent a common chart of accounts.

How it was built

• Supersociedades: statement of financial position, fiscal year 2025. The source file comes in thousands of pesos and the earnings directory in millions; both were converted to current pesos before adding.
• Supersolidaria: financial data of supervised entities, cut-off December 2025, already in pesos.
• Superfinanciera: CUIF by currency, cut-off June 2026, “Total” currency. Profit for the period is taken from account 590000.
• Leverage is liabilities over assets; ROE, profit over equity; ROA, profit over assets. All three are calculated on the regime’s aggregates, not as an average of the entities.

The detail on each universe is in the analyses of the financial sector, the cooperative sector and the corporate X-ray.

All information is public and pertains to legal entities.