Organizational Corruption Differs in Kind from Personal Corruption

There is a claim I have heard made in various forms, and it always arrives sounding worldly and unsentimental: everyone is corrupt to a degree. Everyone rounds a number their own way, everyone weighs their own interest, everyone shades a description in their favour. So there is no fundamental difference between the corruption of a person and the corruption of a corporation or a government — only scale — and singling out organisations is naïve.

This is false, and it is false for three independent reasons. Any one of them would be sufficient.

1. Scale of Consequence

The same quantity of personal corruption produces wildly different effects depending on position.

Misdirect half a per cent of a national budget and the impact on a society is enormous. A private individual cheating by half a per cent of his own budget affects almost nothing.

The obvious reply is that many individuals each cheating slightly would add up to the same total. But that misses the structure: in government, one person acting alone produces the effect of millions. No coordination is required, no conspiracy, no mass participation. A single decision at a single desk.

And this follows directly from what organisations are for. A hierarchy pools the productive capacity of many people to build what none could build alone — aircraft, power stations, road networks. That pooling is the whole point. But it means the people directing the pooled capacity hold power of a completely different order, and their errors and appetites scale with it.

2. Selection: The Promoted Have Been Tested for Corruptibility

This is the argument I find most damaging, and it is rarely made.

How does someone reach the top of an organisation? By climbing. And at every rung, advancement requires demonstrating to those above that you will favour the organisation's interest when it conflicts with the interests of those below you, or with the interests of the wider public.

You are not promoted for costing the organisation money on behalf of outsiders. You are promoted for having shown, reliably, that you will not.

And you demonstrate it repeatedly, which is the crucial part.

Repeated testing sharply increases confidence in a result. A single blood test with a ten per cent error rate leaves substantial doubt; two independent tests give one in a hundred; three give one in a thousand. If an organisation has seven levels, the person at the top has passed the loyalty test seven times, and the organisation is correspondingly confident about what it has selected for.

So the people running large institutions are not a random draw from the population. They are the subset that has iteratively demonstrated willingness to subordinate outside interests — including the public's — to institutional ones.

That is not a claim about their moral character. It is a claim about a filter, and the filter operates whatever the character of the people entering it.

3. The Payoff-to-Risk Ratio Inverts at the Top

I would not become a criminal for a hundred dollars. The risk, however small, is not worth it.

For a million, I might. Not because my ethics changed, but because the pressure did — the amount at stake determines the risk anyone will accept, and that is true of every person including the honest ones.

In corporations and governments, the sums, the influence, and the political power at stake are enormous. So the tolerable risk rises with them.

Meanwhile the risk side barely moves. White-collar offenders are not arrested often, prosecutions are rare, and convictions rarer. So we have benefit multiplied by orders of magnitude against risk that is roughly constant, or lower.

That ratio is what makes the top of an organisation more corrupt, and it would do so even if the people there were drawn at random — which, per the previous argument, they are not.

Why This Matters

The "everyone is a bit corrupt" claim is not merely wrong. It is useful to someone, and it is worth noticing who.

It functions to dissolve the distinction between an individual fiddling an expense claim and an executive making a decision that harms a million people. Once the distinction is gone, so is any special scrutiny of concentrated power — because scrutinising it would be inconsistent, and we all know nobody is perfect.

That conclusion serves precisely the people the scrutiny was for.