Two apparently different things I've seen recently illustrate a common mistake which distorts our thinking about politics and organizations.

First, 30p Lee tried to show that one could eat very cheaply with Tesco's Weetabix. And then when somebody posted a picture of a mouldy Tesla, Musk's incel fanboys claimed the fault was "owner error."

Anderson's mistake is to fail to see that whilst one perhaps could subsist adequately if one were a joyless calculating machine (what a great advert for capitalism), few people are such machines. And Tesla's fans fail to appreciate that even if Teslas do keep well in optimum conditions, few of us are so careful of our vehicles.

The common error here is an inability to see that products and systems must be resilient to human error simply because such errors are widespread and an inevitable part of the human condition.

Risible though these examples are, similar mistakes are common.

Brexiters who complain that Brexit has been poorly implemented should have foreseen the chance that the process would have been undertaken by fallible individuals. In policy-making, implementation must not be a mere after-thought. A policy that only works if everyone is competent is not a serious policy at all.

The same is true of those calling for reform of the NHS. Maybe there is some kind of insurance system which would be an improvement. But this is only a voice in your head. In this world, we must guard against the (high) risk that the task of reforming the NHS will be undertaken by vicious incompetents in hock to US insurance companies.

My call for policy and institutions to be robust to human error is in fact one with a long tradition. Burke famously said:

We are afraid to put men to live and trade each on his own private stock of reason; because we suspect that this stock in each man is small.

Marx said that "the emancipation of the working classes must be conquered by the working classes themselves" because he did not trust individual leaders. (Marx and Burke aren't so unalike).

The US's founding fathers created a system of checks and balances "to control the abuses of government." "If angels were to govern men, neither external nor internal controls on government would be necessary" wrote James Madison, knowing that men were far from angels.

And Warren Buffett has said: "I always invest in companies an idiot could run, because one day one will." He looks for companies that are resilient to human error. The damage caused by such error can be enormous; one cause of the financial crisis was Fred Goodwin's tyrannical management of Nat West.

So, what would a polity that embeds this principle look like?

In some ways, like our present one. There are nine decision-makers on the MPC to help guard against any one of them being fallible. Judges are constrained by sentencing guidelines to avoid rogue ones making erratic decicions. Where markets function well, they eliminate the bad products resulting from human error. And there's increasing use of algorithms because, as Daniel Kahneman and colleagues have concluded, "simple models beat humans" (although of course algos can go wrong because they are made by flawed humans).

But could we do better? Yes. Here are four possibilities:

- A more generous welfare state. In my context, the case for this isn't just that Anderson is wrong and people need an income well above subsistence levels because they cannot budget with maximal efficiency. It's also because fiscal and monetary policy-makers cannot adequately foresee the future which means that discretionary policy is unreliable. Stronger automatic stabilizers - a higher safety net - thus do a better job of protecting is in bad times.

- Improving selection mechanisms. Our media-political institutions select not for the best but for the worst: they attract incompetent MPs and ministers, and filter out intelligent ideas. They thereby magnify our vulnerability to the human factor: the story of Truss and Johnson is one of systemic rather than individual failure. We must think how to reverse this.

- Increasing diversity. Cognitive diversity reduces the danger of us accepting bad policy by ensuring that all views - not just those of narcissists flattering themselves they are "contrarian" - get heard. Institutions that might facilitate this include citizens assemblies and forms of sortition.

- Economic democracy. The UK has a shortage of good managers. Econ 101 says we should economize on scarce resources. We must find ways of reducing our dependence upon incompetent, rent-seeking or hubristic bosses. One way to do so is strong competition policy: in well-functioning markets, bad companies are competed out of business and easily replaced by rivals. Failing that, tougher regulation of financial "services", rail companies or utilities is needed; one lesson of the financial crisis is that our economy was too dependent upon a few flawed individuals. And there's a case for worker democracy too.

I know. All these proposals are vague. I mention them not as a blueprint - there's no point writing a menu if you don't have a kitchen - but to highlight how far short our existing institutions fall from the ideal of ensuring resilience to bad decision-makers.

This issue is, however, off the political agenda. One of the many defects of our political debate is a belief that things will be tolerable if only we could find the right people; this is Bonnie Tylerism, holding out for a hero. This, however, is the wrong question. We should be looking not for good people but for the right institutions, selection mechanisms and processes - devices which would make our economy and politics more resilient to idiots or crooks.