In his book Intuition Pumps and Other Tools for Thinking, Daniel Dennett’s first intuition pump is to extol the virtues of boldly making mistakes. He describes it in this way:
This reminds me about the old joke about how one should go about carving an elephant out of granite. The tongue-in-cheek advice is to just start with a big hunk of granite, and then chip away all the parts that don’t look like an elephant. Dennett is advocating a similar approach – throw the unformed granite block of your thoughts out there, and let all the parts that don’t work get chipped away as your mistakes are found and corrected.
Dennett cheerfully describes himself as “an experienced mistake-maker myself. I’ve made some dillies, and hope to make a lot more.” But there is a key to being a good mistake-maker, Dennett says, and that is a willingness, even eagerness, to let your ideas get chipped away:
Mistakes are good when they are quickly identified, corrected, and learned from. I’m on the record as believing that most new ideas are terrible – and this is true whether the new ideas are coming from private actors in the market, or from policymakers acting from the state. The crucial difference is that in markets, the identification and correction of mistakes is swift, but the same can’t be said for the mistakes of policymakers.
For example, in my post explaining that most new ideas are terrible, I used the example of an upcoming tech product called the R1 Rabbit, and why I thought it would be a flop. Since I wrote that, the product has been released, and the consensus that has developed since then is that the product is, indeed, pretty terrible. Another product I could have mentioned in that post is the Humane AI pin, which seemed even more half-baked to me. This, too, seems to have become the general consensus, and at present Humane is seeing the product returned at a faster rate than they can sell it.
I suspect both of these companies are not long for this world. A significant amount of time, effort, and money was put into building the companies and producing the products. And some people ended up spending money to get a product that turned out to be half-baked. This is certainly not ideal – but the correction is happening swiftly.
Compare this to another bad policy I recently highlighted, when “King William III instigated a tax on windows, on the assumption that dwellings and buildings with lots of windows were likely to be owned by the wealthy, and thus this would serve as a way to tax the rich.” But, as recorded in Scott Hodge’s book Taxocracy, “the tax ‘led to especially wretched conditions for the poor in the cities, as landlords blocked up windows and constructed tenements without adequate light and ventilation.’ Some buildings were constructed with no windows on some floors leading to the ‘propagation of numerous diseases such as dysentery, gangrene, and typhus.’”
This, too, was less than ideal. It caused people to live in unnecessarily miserable, dark, and stuffy conditions. Disease spread more rapidly, costing many people their lives and inflicting significant pain and suffering on those who survived. This, too, was eventually fixed – the tax was eventually repealed. But the tax and its ill effects lasted for 150 years before that happened.
Entrepreneurs are not necessarily smarter than state policymakers, nor do they have an intrinsic ability to come up with better ideas. But when entrepreneurs make mistakes, the error doesn’t last long. Policymakers can make mistakes that cause disease, suffering, and death and those policies can continue on for multiple human lifetimes before the error is corrected.
Arnold Kling often used the mantra “Markets fail. Use markets.” Yes – because when markets fail, they fail swiftly and correct swiftly. Dennett advises his students to guard against hiding their mistakes, especially from themselves. But state policymakers have the ability to hide away or overlook their mistakes in a way that simply doesn’t exist in the market, making those mistakes so long-lived as to border on immortality.
READER COMMENTS
Richard W Fulmer
Aug 16 2024 at 11:11am
Fail fast and fail cheap. Computer simulations help achieve to both goals.
Wealth is useful too. Primitive people living on the edge of starvation, where error often means death, can ill afford trial-and-error experimentation. Ancient peoples were, therefore, probably very conservative – suspicious of change and of those suggesting it.
Small populations also limit innovation – fewer people mean fewer ideas. So, humanity had to reach a critical mass of both population and sufficient cumulative innovation so that new ideas were both available and no longer viewed as existential threats.
Craig
Aug 16 2024 at 11:20am
“But when entrepreneurs make mistakes, the error doesn’t last long.”
Indeed, one exception is continuing to pursue the marginal business.
Warren Platts
Aug 16 2024 at 12:49pm
I think Dennett is about my favorite modern philosopher. Certainly top five!
Monte
Aug 16 2024 at 1:11pm
Markets are indifferent to good intentions (Fiat justitia, ruat caelum!), where governments thrive on them. Like embedded Rube Goldberg machines, interventions are often long-term, over-complicated, and inefficient. Milton Friedman said it best:
Thomas L Hutcheson
Aug 16 2024 at 5:43pm
This is the way I feel about taxation of net CO2 emissions (or merit based immigration, or progressive consumption taxes, congestion taxes). Sure as we get better at modeling the effects of CO2 accumulation, eliminating other policies aiming at net reduction of CO2 emissions, and forming clearer ideas of how rapidly new technologies reduce the cost of avoiding CO2 emissions, the calculated optimal rate of the tax on net emissions will change. But we can’t reform what does not exist.
nobody.really
Aug 16 2024 at 6:33pm
Release of human waste into the water supply leading to cholera and typhoid
Overfarming and irrigation leading to aquafer depletion and (in the Futile Crescent) desertification
Slavery leading to … slavery
Trains and power lines leading to wildfires
Lead in paint, and in gasoline leading to lead pollution
Sulphur dioxide emissions leading to acid rain
Carbon dioxide and methane emissions leading to climate change
Plastics, pharmaceuticals, and PFAS ended up in the water supply, leading to various problems
We shouldn’t be blithe about the long-term harms of private actions, too.
steve
Aug 17 2024 at 5:50pm
It’s actually pretty easy to think of cases where the actions of private entities were harmful and persisted for a long time. It’s also easy to think of instances where governments quickly corrected mistakes. Incentives do matter and someone who wants to stay in office or a government worker who wants to keep their job will need to correct mistakes. Its believable that private incentives might be stronger but I dont know if there has ever been an empirical look at this.
Steve
Kevin Corcoran
Aug 18 2024 at 5:16pm
Sure, but like I said to nobody.really, my claim was not that “private entities” never engage in “harmful” actions over “a long time.” My claim was that mistakes are quickly identified and corrected when they take place in the context of market activity. Maybe you are defining “market activity” to just mean any “actions of private entities,” but that would certainly be an…odd definition.
The idea that elected officials who want to stay in office are therefore incentivized to identify and correct mistakes only works if voters are able to consistently and accurately identify those mistakes, know which elected officials are responsible for them, to what degree, and use this information to inform their votes. But the evidence against that is pretty overwhelming. You yourself have frequently commented that as far as you can see politics is all tribal, people just vote for whoever is on “their team,” etc.
There has actually been a massive body of empirical research on this question over many decades. If it would be of interest, you can get a free pdf copy of a book by the center-left economist Cliff Winston of the Brookings institute at this link, which empirically compares market failure with government failure. But the short version – in almost all cases, government failure is empirically worse than market failure, and in practice government attempts to correct market failure almost always end up being worse than nonintervention. Not always, but in the substantial majority of cases.
Kevin Corcoran
Aug 18 2024 at 5:00pm
Sure, those are certainly cases of bad things happening, but you seem to have missed the point. My claim was not that private actors never do bad things over extended periods of time. What I actually said was about how “in markets, the identification and correction of mistakes is swift.” And obviously, not all activity that private citizens or organizations engage in is market activity. Most of the examples you list are cases of pollution. Steven Landsburg once argued that in the vast majority of cases when people speak of “market failure,” what is actually going on is a case of market absence. That is, there is a “market failure” regarding things like air and water pollution because there is no market for such things – nobody has clearly defined property rights regarding air and water, transactions costs are too high, among other problems. So, yes, private citizens or companies (and governments too!) may end up dumping toxic substances into waterways over an extended period of time, and yes, that is bad. But this does not change the point I was making, which is that in markets mistakes are quickly identified and swiftly corrected. Externalities, by definition, are actions that are not encompassed by market activity. That’s the very reason why externalities are usually considered troublesome! The proverbial “tragedy of the commons” is also a similar case – a commons, by definition, is a place where nobody has any property rights, meaning that counterproductive actions can carry on over an extended period precisely because a commons is not priced in the market. So pointing to a variety of behaviors that went on for a long time without being corrected but were also taking place outside of the market doesn’t strike me as a very interesting response to what I actually said.
You also brought up slavery as an example. I assumed you would already understand that when I (and others on this site) talk about market activity, we are already taking for granted a generally libertarian or classical liberal framework that presupposes slavery is an illegitimate institution and that markets in slaves are also illegitimate. But if you need that to be explicitly spelled out going forward, I can make an effort to re-emphasize that point in future posts.
nobody.really
Aug 19 2024 at 1:46am
1: My kid complains that her medicine doesn’t taste as good as ice cream. Yet medicine may be designed to cure her illness, whereas ice cream is not—thus, even if everyone agreed with her claim, I don’t know what difference it would make, since she’d still need to take the medicine.
Kevin Corcoran claims that market transactions have advantages that government actions lack. Yet government actions may be designed to address certain things that market transactions are not—such as managing externalities or providing public finance—thus, even if everyone agreed with his claim, I don’t know what difference it would make since we’d still need government actions.
That said, [insert discussion of spontaneous order for managing externalities/common pool problems here].
2: Is slavery a sui generis example, or is Corcoran’s objection just a No True Scotsman defense? I expect some vegans regard ownership of any kind of animal as akin to slavery. Can we dismiss those concerns by simply averring to the norms of classical liberalism? I think Corcoran is trying to make a point about the dynamics of markets, but I find the norms against slavery to be only tangentially related to those dynamics. I don’t see why the laws of supply and demand would not apply to slaves, moral qualms notwithstanding.
3: More generally, I understand markets to promote efficiency. If you have moral qualms about anything promoted by a market, you will likely find that the market more efficiently promotes those things you have qualms about.
Slavery dates from time immemorial—or earlier. (The new lion kills the old, and then kills the old lion’s offspring despite the opposition of the lionesses, so that the new lion can monopolize their reproduction. The fact that the new lion commandeers the reproduction of the females against their will looks slavery-like to me.) Still, the existence of markets made slavery more efficient for slave-owners, and made investments in slaves more liquid.
Some people have qualms about prostitution. Arguably the ability for people to market themselves online as escorts has enhanced the prostitution business.
Many people suffer from weak impulse control. They may find themselves addicted to chemicals, or sex, or gambling. Markets crop up to cater to their demand, even when doing so is self-destructive. The Babylonians and Egyptians described the harms of alcoholism—and we continue to observe these harms today. Can we call this a long-term adverse consequence of a market activity?
4: Finally, I don’t know if we can dismiss all my examples as externalities. The people who bought and used lead-based paints and gasoline may have externalized some of the consequences of their decisions to their neighbors, but they internalized a lot, too—in every sense. As far as I know, these harms arose from good-faith ignorance about the consequences of lead. But nothing about the market nature of the transactions inoculated people from the consequences, or shortened its duration–at least, as far as I know.
Kevin Corcoran
Aug 19 2024 at 10:06am
Time for another of my semi-regular reminders to you that I prefer to be addressed by my first rather than last name 😛
That aside, just a few quick thoughts before I have to turn my attention to Monday morning, in all its glory.
You write:
The key point is that in many (not all, but many) cases, what we actually have is not government actions occurring where market transactions can’t take place. Instead, we get government actions that forbid market transactions from taking place where they are easily able to do so, and requiring government actions instead. The list of areas where moderate, mainstream economics suggests government can usefully intervene is a tiny in comparison to what the government actually does.
So, even if we grant your claim that “we’d still need government actions” to account for “certain things market transactions are not” able to address, that still undercuts about 90% of what the government actually does. And as Bryan Caplan once pointed out, “market failure theory is also a reproach to every existing government. How so? Because market failure theory recommends specific government policies – and actually-existing governments rarely adopt anything like them.” That is to say, in the real world, when market failure actually exists, actual government policies that are supposed to address them tend to be either costly and useless, or outright counterproductive.
(In a way, this is similar to the thought I have when people say things like “Even Adam Smith didn’t think regulation was always bad – look at these various cases where Adam Smith advocated this or that intervention!” Yet, if you take the sum total list of regulations and interventions about which Smith spoke favorably, the totality of that list is pretty trivial, especially in regards to what actual policy currently looks like. If we had a government that intervened in each and every way Adam Smith is said to support, we’d reduce the size of the government to a miniscule fraction of what it actually is. That’s before considering that half the time, the alleged support from Smith turns out to be a half-quote where the full context makes it clear he’s not actually in favor of said intervention.)
The No True Scotsman fallacy is when you try to say X isn’t really Z, because X has (or lacks) some characteristic that is neither required for or incompatible with being Z. So saying “No true Scotsman mixes his whiskey with water” is a fallacy because it arbitrarily requires that an inessential characteristic (water in whiskey) overrule an essential trait (one’s lineage). What I had said was that my posts on this blog are written from within a classical liberal/libertarian perspective. And it seems to me that an opposition to slavery is an essential characteristic of being a classical liberal/libertarian. Maybe you have a different understanding of what liberalism entails.
Yes, this is true. I am myself vegan, and I believe that factory farming is a great evil. (Though I don’t believe that animal ownership as such is akin to slavery, for what that’s worth.) But I also can’t help but notice that in practice, the government massively subsidizes factory farming to the tune of something like $20 billion a year. Maybe some would say my classical liberal beliefs mean I shouldn’t support government restrictions on meat eating – but I think I can at least reasonably think government should stop pouring massive amounts of gasoline on the fire.
You also wrote:
Indeed, Jeremy Black’s book Slavery: A New Global History was by far one of the better books I’ve read in learning that lesson. I was, unfortunately, a victim of the American public school system, and among the lessons I was taught in school was that slavery was an exclusively American institution, which consisted entirely of white American’s kidnapping black Africans and enslaving them. The fact that, for example, slavery existed in Africa long before Europeans ever arrived there, that African slaves were sold to Americans and Europeans primarily by other Africans who were themselves slave owners, and that slavery persisted in Africa long after it was banned in Europe and America, was never mentioned. I didn’t learn about any of that until I was well past high school and just did some reading on my own.
But you also wrote:
To put it as kindly as I can, this particular claim is, shall we say, wildly contested. Many historians have written many books arguing that societies becoming more and more market oriented weakened rather than strengthened slavery as an institution and hastened its end. Maybe all those arguments are wrong, but I don’t think you can simply assert otherwise.
(Lengthy comment, I know, but for me this is what “quick thoughts” looks like. I am very much the kind of person characterized by the old saw “If I’d had more time, I’d have written a shorter letter”)
nobody.really
Aug 19 2024 at 2:07pm
Wait a minute. Look, I was ok with the whole “preferred pronoun” thing. But now we’ve got to deal with preferred NOUNS too? Where will it end? It’s bad enough when I’m talking to all these nobles and their preferred ADJECTIVES (“Your Highness,” “Your Excellency,” “Your Grace,” etc.) Next time I’m gonna respond with my own preferred interjections.
Yeah, Kevin. Sorry ’bout that.
Regarding teaching about slavery: According to the Online Etymology Dictionary, the word “slave” derives from the word “Slav.” That is, the archetypical slave was a Slav–someone with fairly white complection. For what it’s worth.
john hare
Aug 16 2024 at 7:01pm
One of the skill sets required of an inventor is the ability to drop ideas that aren’t working. Other wise one spends far too much time and resources chasing bad ideas. Stay the course sounds good, and is proper when correct, but can easily lead to target fixation on the wrong target.
I’m in the middle of a product development now that I hope will be the good one. But if I am wrong, I haven’t quit my day job.
David Seltzer
Aug 17 2024 at 2:53pm
Kevin: Nice post. I tutored AP calculus. My counsel to students, mistakes are your mentors. You will learn from them. I asked some why they feared mistakes on diagnostic tests that are designed to spot weaknesses they could address. Often, it was because they feared they weren’t as smart as they thought they were. In some cases, their confidence was a bit shattered. In the military, I heard “Progress! Not perfection! That aphorism has been very useful.
Knut P. Heen
Aug 20 2024 at 5:52am
The definition of a long-term investment is a short-term investment that failed. It happens in the stock market too.
Comments are closed.