In short

There are four, and treating them as one thing called risk is why lenders mitigate the wrong one. Default: the borrower does not pay. Illiquidity: your money cannot come back early. Concentration: too much of it is in one place. Capital at risk: the downside is the money itself, not a fall in value.

The fourth is the asymmetry. Your best case is the rate you agreed; your worst case is everything you lent. Those are not opposite ends of one scale, and nothing about a single loan makes them so.

01Four separate things

The riskWhat happensWhat actually reduces it
DefaultThe borrower does not pay, on time or at allAssessment before you lend, and security that survives the arithmetic
IlliquidityYou need the money and cannot have itNot lending money you might need. Nothing else works.
ConcentrationOne failure takes a disproportionate share of your capitalMore loans, smaller each. Only this.
Capital at riskThe loss is the capital, not a drop in a quoted valueAccepting it, pricing it, and sizing each loan accordingly

Notice that security appears once, against one of the four. It does nothing at all about the other three — which is the whole problem with the phrase it is secured, so it is safe.

02What one total loss actually costs

This calculation is the single most useful thing in part three, and it takes a minute.

R1 million lent at 18%, split different ways, with one borrower failing completely. Illustrative arithmetic, not a rate anybody is offering.
LoansEachAt 18%One total lossResult
2R500 000R180 000−R500 000−R320 000. Two years of gains gone
5R200 000R180 000−R200 000−R20 000. A lost year
10R100 000R180 000−R100 000+R80 000. A poor year, survived
20R50 000R180 000−R50 000+R130 000. Barely noticed

Same capital, same rate, same single failure. The only variable is how many pieces it was in — and it is the difference between a rounding error and a disaster.

03One loan is not a portfolio

A rate compensates for default risk across enough loans for the arithmetic to work. That is what a rate is. It is an average price for an average outcome over a population.

On a single loan there is no population. There are two outcomes: you are paid, or you are not. The rate was never a prediction about your one borrower, and it cannot be.

This is also why institutional private credit performance figures should never be quoted in support of one private loan. Those returns were produced with hundreds of loans, a credit team and negotiated covenants. The asset class is the same; the defences are not.

04The concentrations that do not look like concentrations

A book can look diversified and be concentrated in ways that only appear when something goes wrong.

KindWhat it looks likeWhat breaks it
BorrowerSeveral loans to the same person or groupOne borrower fails and takes several loans
SectorTen loans, all to property developersRates move or a market stalls, and all ten are hit at once
GeographicEverything secured in one townOne local event hits every piece of security together
TimingEverything falling due in the same quarterAll your money is idle at once, or all at risk at once

Sector concentration is the one that catches experienced lenders. Ten different borrowers is diversified against borrower failure. It is not diversified at all against the thing that makes all ten fail together. Related borrowers — a group, a family, companies with the same director — count as one.

05The asymmetry, stated plainly

It is worth writing the two numbers next to each other once, in your own handwriting.

  • Best case: the interest you agreed. A number you can calculate exactly today.
  • Worst case: the capital, plus the cost of trying to recover it, plus the time.

A private loan is therefore a bet in which you know the size of the win and the size of the loss, and the loss is larger. That is not an argument against lending. It is an argument for sizing each loan so that the loss is survivable — which is a decision you make before you ever meet a borrower.

Next: the risk that has nothing to do with the borrower at all.