It can now be said with certainty that luck matters more than talent and effort. Not that talent and effort don’t matter, but that they are easily overwhelmed by bad luck, and easily replaced by good luck. Consequently, all ideologies that depend on any version of Just World Theory (as, for example, most versions of Libertarianism and Anarchocapitalism do) are false and must be abandoned. Equality of opportunity requires considerable adjustments be made for the unequal distribution of luck in any given population. And more humility is warranted from the successful, and more empathy for the disadvantaged, than most are naturally inclined to. The rich are not rich because they are better or tried harder. The poor are not poor because they are inept or lazy.

Here I will discuss some new science and data that not only make this inescapably clear, but illustrate empirically how that actually changes everything we should think about how to organize our mutual society. Conclusion? Limited redistribution policies like a Universal Basic Income actually make a lot more sense than you might have thought.

Atheism vs. Ideology

A political philosophy is eclectic and revisable, allowing evidence to overthrow assumptions. A political ideology, by contrast, is purist and dogmatic, allowing assumptions to overthrow evidence. And it is political ideologies that have now become the new religions, built on top of myths, dogmas, faith beliefs, ignorance, delusions, and fallacies (a recent example: the various latent, and sometimes explicit, ideologies embraced by fans of Jordan Peterson). And just like the religions of old, false ideologies always sustain themselves on mythologies built out of half truths. For example, using the fallacies of selection bias and false generalization, actually cheating or lazy or dumb poor people will be held up to “prove” that this is why all poor people are poor; and honorable, hard working, smart rich people will be held up to “prove” that this is why all rich people are rich. Never mind the vast statistical data refuting both generalizations. There are a disturbing number of corrupt, lazy, stupid rich people, and honest, smart hard working poor people. This should worry you. A just social system should not tolerate this outcome. And the system that actually will minimize this outcome the most, and most efficiently, cannot be discovered from the armchair. It cannot follow from ideology at all. It can only be discovered empirically.

Indeed, this is the entire point of atheists complaining about God’s injustice as evidence against his existence, and the need of humans to step up and fix what the absence of God saddled us with. No God polices or judges our criminals; so we have to. No God saves those ruined by disaster; so we have to. No God prevents floods, fights fires, educates kids, cures disease or heals wounds, stops earthquakes from toppling bridges and buildings; so we have to. We know there cannot be a just God because if there were any God at all, then he set the world up unfairly, distributing good and bad luck without regard for who deserves either. But that entails believing that a just God would set the world up fairly. And that entails a society cannot be just, if it acts just like an unjust God; and can only be just, if it does what a just God would do.

Which still of course means, “within its means,” as a society is not literally a God; it lacks the tremendous advantages of infallibility and unlimited resources. But the parallel exists on a continuum: as a society gains a surplus of resources, through the accumulating of wealth, and reduces its fallibility, through improved systems of auditing, testing, and control (the checks and balances enshrined in the U.S. Constitution being merely one example of doing that, in contrast to a monarchy or autocracy), then the nearer it gets to the capabilities of a God, the more justice it should therefore be capable of enacting—that any actual God failed to enact. What God did not do—because God does not exist—we must step up to do, once we are safely able to. Which then always becomes a debate as to what we are safely able to do. And that, too, can only rationally be adjudicated with facts and logic, rather than emotion and mythology.

I’ve discussed before how empirical evidence and analysis continually changes and updates my political opinions—and how I’ve become an eclectic hybridist with respect to political theory. As I wrote in my 2016 Revision Notions for my 2005 book Sense and Goodness without God:

I have … changed many policy positions in the Political Philosophy section, after having studied far more research and evidence pertaining in the decade since I wrote. For example, my perspective on the economics of tax regimes (e.g. p. 394) has changed completely.

For instance, it turns out, in most cases an excess of transactional taxes (tariffs, sales tax, etc.) becomes an unjust mechanism for generating collective revenue, because they disproportionately harm the poor and unfortunate (and also reduce economic activity, which reduces the overall wealth of a society), whereas progressive income taxes are a far more efficient and justifiable mechanism for tolling a hoarded surplus (merely changing what money gets spent on, not how much money transactions cost). State commerce could still be a useful revenue source, but cannot replace taxing the rich, as again any commerce’s revenues arise mostly from the poor, not the wealthy. The wealthy always just skim off the top of everyone else’s transactions. We should therefore skim off the top of what they take, and recycle it back into public goods.

Likewise:

I also see the merit of emphasizing more the need of hybrid socialist-capitalist regimes (each a check and balance against the other, with a strong equilibrium producing the most sustainable societies) and that includes more attention to national healthcare systems and examining the possible merits of universal basic income (which I would link as a lifetime reward for a stint of national service, including peace corps and other nonmilitary endeavors).

That last a point I will explore more here in my conclusion below. But the overall point is that unchecked capitalism leads to as much ruination, injustice, corruption, dysfunction, and failure as unchecked socialism (which, when so unchecked, we often call “communism”). We need socialism as a check and balance against excesses and harms of capitalism, just as we need capitalism as a check and balance against the excesses and harms of socialism. The most sustainable, just, and prosperous society lies at the equilibrium state between those two counteracting forces. Leaving the only question: Where exactly, among all the optional ways of setting up a society, is that equilibrium state?

And finally:

I did warn in the opening of the Politics Part that my conclusions there were the most likely to change (e.g. p. 369). That was accurate prophecy. I have some more refined insights now even on the grounding of political theory. For example, there really is no objective difference between public sector and private: governments, like unions, are just another corporation, with their own bylaws and shareholding structure; and all the problems that plague the public sector plague the private sector as well, and vice versa, with no practical distinction in reality; and the solutions are likewise often the same. Much more could be said. But for now, just note I still agree this is the section where the most disagreement is possible; and I might no longer hold a position you find there. This was 2005, well over a decade ago. A lot has changed since then.

Quite so. And that continues to be the case. What remains the same is the basic political epistemological and axiological framework I discovered and lay out in Sense and Goodness: all conclusions in the political domain must be based on evidence, not ideology (what actually works and doesn’t can only be ascertained empirically; outcome measures determine all truths); and the only defensible goals of any political system are (1) the regulation of the use of power and (2) the sustaining of a civil society, as being the only condition that can afford to be permissive of more basic freedoms.

You can see what I’ve said on political subjects outside Sense and Goodness on my blog. Most of what’s on my old blog is obsolete now, except perhaps my Factual Politics series. Otherwise, as of this writing my most informative articles are:

• Sic Semper Regulationes
• Money Buys Happiness? Not After You Hit Six Figures
• On School Nutrition: How Republicans Create an Alternative Reality
• The Shocking Reasons Why We Should Go Nuclear
• Myths of Charity: The Enduring Sham of Arthur Brooks
• Criminalizing Pot … a Plot to Punish Liberals & Minorities
• Sexy Sex Sex!! (for Cash on the Barrel!)

As you will see across all these articles, in every case, evidence is always the deciding factor, trumping myths and armchair assumptions—and ideology—every time. And the only worthy goals we find in the end are a civil society that limits abuses of power by its citizens. Which requires empowering a government to maintain those limits. Because an excess of freedom, always leads to an excess of abuses. Only restraint prevents abuse. So restraints must exist on everyone—including the restrainers. Leaving the only question: What restraints do we need upon the restrainers, to allow them to restrain abuses of power from individuals (every crime being itself an abuse of power, whether claimed and wielded by thieves, fraudsters, embezzlers, murderers, or anyone the like), without themselves abusing the power we gave them to do that?

One thing that becomes clear, is that you can’t have a civil or just or prosperous society unless you pay for it. Leaving the question: How do we pay for it? Who pays? And how much? As with every other question above, the answer must come from evidence, not ideology.

The Italian Study

Scott Kaufman summarizes in Scientific American a lot of recent economic science on this question (I highly recommend reading everything he links to and mentions), including a brilliant computer modeling study recently completed in Italy. See “The Role of Luck in Life Success Is Far Greater Than We Realized” (1 March 2018). I am usually very skeptical of computer modeling studies; after all, how empirical can they be, it’s not reality in the computer? But this is an exception. Because it operates on principles of simplicity similar to those worked up in the field of Game Theory, where the models created actually do track to reality, there being a one-to-one correspondence between the premises in the model, and the facts on the ground. And they are testable. (See my discussion of Game Theoretic models and their empirical uses for moral theory in The Real Basis of a Moral World.)

As Kaufman describes it:

The Italian physicists Alessandro Pluchino and Andrea Raspisarda teamed up with the Italian economist Alessio Biondo to make the first ever attempt to quantify the role of luck and talent in successful careers. In their prior work, they warned against a “naive meritocracy”, in which people actually fail to give honors and rewards to the most competent people because of their underestimation of the role of randomness among the determinants of success. To formally capture this phenomenon, they proposed a “toy mathematical model” that simulated the evolution of careers of a collective population over a worklife of 40 years (from age 20-60).

They reduced everything to three simple measures:

• Generic units of success, or “resources,” beginning with a starting amount (thus eliminating any need to be particular over what we are measuring as success or counting as resources; could be money, doesn’t matter, though mostly money is the intended analog);
• A generic measure of “talent,” which is simply defined as a degree to which one benefits from encountering opportunities (thus eliminating any need to be particular over what talents we are talking about, could be “intelligence, skill, motivation, determination, creative thinking, emotional intelligence,” doesn’t matter; a single measure is simply the aggregate effect of all of one’s talents, whatever they are);
• And how frequently bad or good luck “events” occur in a person’s life—bad luck always reduces their success, while good luck, always increases their success in proportion to “talent” (thus eliminating any need to be particular over what events or opportunities or advantages we are talking about, it’s just anything whatever that, if it happens to you, will set you back, or present to you an opportunity you might be able to leverage into getting ahead).

This is a reliable model in construct because each generic measure does in fact track to reality. All opportunities in the real world, of whatever kind, do indeed merely reduce to what you can leverage to get ahead; all setbacks in the real world, of whatever kind, do indeed merely reduce your resources or access to resources; and all talents in the real world, of whatever kind, do indeed merely sum to an aggregate ability to leverage resources and opportunities into success.

One need merely tweak these three measures to align the model with their distributions in reality (different distributions of talent; different distributions of starting success; different distributions of setbacks and opportunities). But if whole ranges of distributions produce the same effect, you don’t need to test alignment with reality. For example, if no matter how you distribute talents, the same outcomes result (or the same outcome tendencies—which is what, actually, we are looking for), then those will be the outcomes we can expect in reality, because there is literally no other way to arrange a system. And even if this is only true for certain ranges of distributions, we can then check to see which ranges more closely fit reality. For instance, if we give the same exact average talent to everyone and a high talent to a scant few: we can already observe, factually, that is not the distribution in reality; so we don’t need to model it.

This is clever. And the advantage this gives us is that we can test a million different models, without the unethical and ungodly expense of running human experiments. If we worry the observed outcomes of a model might be the result of unrealistic distributions of the variables, all we need do is change those variables to see if, indeed, different outcomes are even possible. The rest is either analytical fact (all distributions have the same outcome) or empirical fact (those distributions that have a different outcome we can then go and observe either do or don’t align with actual distributions in reality). The model’s outcomes can also be directly tested in reality. But more on that in a moment.

What Their Simulations Prove

What Pluchino, Biondo, and Raspisarda (or PBR) found was that no matter how you tweak the models (within any bounds of plausible realism), the same results kept occurring. Unless you change how you distribute resources. Any realistic talent distribution? General outcome is always the same for any given resource distribution. Any realistic luck distribution? General outcome is always the same for any given resource distribution. There are variances in absolute values in the outcomes, of course; but the overall shape of the distribution among those outcomes does not meaningfully differ.

Thus, resource distribution is the only relevant variable. You can try to change the distribution of talents, for instance (let’s say, by genetic engineering, or free universal education or nutrition or something), and that could help. But it still won’t override the effects of the resource distribution. You can try to change the distribution of luck, for instance (let’s say, by making the environment “safer” for everyone), and that could help. But it still won’t override the effects of the resource distribution. And really, even those things are just more resource distribution—genetic engineering or access to education or nutrition is just another “opportunity” that will be leveraged by talent; making the environment safer is also just another “resource” that will be leveraged by talent. So trying to make everyone more talented or lucky is simply analytically identical to giving the same resources to everyone. This is a fundamental insight their model demonstrates.

For example, the PBR models showed that no matter what you do, leaving the system to “The Invisible Hand” as proposed by the original philosopher of capitalism (or so people imagine), Adam Smith, always generates the same overall, and rather odd, result: the people who end up on top (“the rich”) are disproportionately merely mediocre in talent; while most of the talented people end up stranded in the middle, or crushed at the bottom, reduced to poverty—indeed, more often they end up even poorer than they started! Thus, statistically, capitalism mostly just rewards luck and punishes talent. This is quite counterintuitive, but when you examine why this happens, it becomes easier to grasp: once any misfortune befalls you, it becomes increasingly more difficult to catch back up; whereas if you start with more good luck, you quickly accumulate enough resources to easily weather numerous misfortunes. The lucky thus become invulnerable while the unlucky get mowed over; talent, makes next to no difference. This is why it is almost impossible to escape poverty, no matter how hard you work or how talented you are; and almost impossible for a rich person to become poor, no matter how incompetent or lazy they are.

(We’ve well known this. Linda Tirado, “Why Poor People Stay Poor” at Slate illustrates how misfortune keeps even hardworking, talented people stuck in the churn, unable to escape. That article’s an excerpt from her equally relevant book, Hand to Mouth: Living in Bootstrap America, which supplements Ehrenreich’s Nickel and Dimed: On (Not) Getting by in America. Both provide ample factual grounding for the principle, and what it would take to fix it. There’s also a lot we could say about lazy, useless rich people.)

As Kaufman describes it:

So what did the simulation find? On the one hand, talent wasn’t irrelevant to success. In general, those with greater talent had a higher probability of increasing their success by exploiting the possibilities offered by luck. Also, the most successful agents were mostly at least average in talent. So talent mattered. However, talent was definitely not sufficient because the most talented individuals were rarely the most successful. In general, mediocre-but-lucky people were much more successful than more-talented-but-unlucky individuals. The most successful agents tended to be those who were only slightly above average in talent but with a lot of luck in their lives.

And this is confirmed in the real world. For instance, Kaufman describes an empirical study that “looked at whether larger grants lead to larger discoveries” in the sciences. The results matched the PBR model:

They found a positive, but only very small relationship between funding and impact (as measured by four indices relating to scientific publications). What’s more, those who received a second grant were not more productive than those who only received a first grant, and impact was generally a decelerating function of funding. The authors suggest that funding strategies that focus more on targeting diversity than “excellence” are likely to be more productive to society.

In other words, by choosing to reward scientists who are “lucky,” on the false assumption that they weren’t lucky but somehow better scientists, we find out our assumption was wrong: they perform no better than other scientists. Whereas if we gave more and smaller rewards to more scientists, disregarding past success, we will get more results leveraged from “luck,” and thus more scientific success overall. It is counterintuitive—but clearly correct. This is predicted by the PBR model. And observed in practice.

This is a phenomenon actually already well understood by experienced investors: if you just fund one business with a ton of money, in the hopes of leveraging profit from it, you are highly prone to losing your shirt, because the average rate of business failure simply becomes your probability of losing everything; whereas if you fund ten businesses with a tenth of that same money each, you will get ahead, even as several of those businesses fail, since then the average rate of business success simply becomes your return on investment. It is thus again counterintuitive: you have to invest in failure to increase your probability of success.

Fact thus trumps intuition. For example, you will have ideologues who hold up the failure of the solar panel manufacturer Solyndra as a reason the government shouldn’t “pick winners and losers” with things like loan programs, while ignoring that in fact this is what all investors do, because as evidence proves out, the net effect of the government’s investments can only be positive if several plays are bet; we expect to lose some—because that’s the only way we win some. Ideologues thus hold up Solyndra as proof of their ideology, by ignoring all the companies funded by the same program that didn’t fail. The government is now making a profit on that program. Reality trumps ideology.

(I’m setting aside the other fact such critics ignore, that there was a national security reason for that energy loan program, which we’re now noticing, as our dependence on China for essential goods makes us extremely vulnerable; people likewise forget the importance of public works programs in forestalling a national economic collapse. That’s not germane to my present point. But it’s relevant to anyone concerned with that issue on its own. That government loan program was not created on a lark. And it’s expanded function after the recession of 2008 was not far different from our economic business loan stimulus program today.)

So what happens when you change how resources get distributed? Well, they tried every possible way of doing that, ran millions of sims, and aggregated the statistical results, charting “the most efficient funding strategies over the 40 year period in descending order of efficiency,” efficiency meaning, “the least amount of funding for the greatest return on the investment.” The outcome is fascinating:

Starting at the bottom of the list, you can see that the least effective funding strategies are those that give a certain percentage of the funding to only the already most successful individuals. The “mixed” strategies that combine giving a certain percentage to the most successful people and equally distributing the rest is a bit more effective, and distributing funds at random is even more efficient.

In fact:

The best funding strategy of them all was one where an equal number of funding was distributed to everyone. Distributing funds at a rate of 1 unit every five years resulted in 60% of the most talented individuals having a greater than average level of success, and distributing funds at a rate of 5 units every five years resulted in 100% of the most talented individuals having an impact!

In other words, just as with wise business investing and anything else, by investing in failure, we greatly magnify success: funding untalented individuals equally with talented individuals, results in more talented individuals becoming successful. By a rather enormous amount in fact: from almost all talent failing (in the “ufettered capitalist” model), to all talent succeeding (in what we might here call the “universal basic income” model, or UBI model).

And this result comes out no matter how you tweak the models. There is simply no other way to create a society that has most talent succeed. All other models grind the talented into the floor, or leave them floundering in the middle, and mostly only make mediocre people rich (which in effect, hands over all state power to them as well—for we all well know there is a direct correlation between wealth and power). There is no possible way this is a good idea. Such a system therefore cannot be described as in any way good, just, rational, or even desirable. (Not even to the mediocre, I might add: as per John Rawls, only a lucky few of them get the boon; so even the mediocre should not want this system. And indeed, PBR modeling showed more mediocre people by far benefit from a UBI system than pure capitalism.)

Another verification of the PBR model is in IQ correlation studies. As reported in 2007 by Jay Zagorsky in “Do You Have to Be Smart to Be Rich? The Impact of IQ on Wealth, Income and Financial Distress” (Intelligence 35), there is no causal correlation at all between IQ and wealth, and though there is a correlation between IQ and annual income, that correlation is actually rather small and flat. These are not what we would expect—but for the PBR thesis.

As to wealth, rich people simply aren’t that much smarter than poor people. Indeed, once you control for things like “having been raised in a wealthy household,” there is no statistically significant correlation remaining between IQ and wealth. Such factors are instead known to be the cause of a higher IQ (rather than a high IQ causing them). And yet even if we don’t control for things like that, the raw correlation between IQ and wealth is still surprisingly small. As Zagorsky found, “people with above-average IQ scores are only 1.2 times as likely as individuals with below-average IQ scores to have a comparatively high net worth,” which means, “relatively large numbers” of people with low IQs are rich. And even to the extent that there are more rich people with high IQs than poor, this is entirely explained by luck, not talent: rich people are only that 1.2 times more likely to be smarter insofar as they were advantaged to develop more of their potential IQ by the fortunes of their environment (like “growing up rich”). Once you control for all that, no correlation remains.

And as to income, here is the graphed correlation in Zagorsky:

Instead of a 1:1 correspondence (which would put most of the dots in the bottom right quadrant into the top right quadrant), high IQ barely helps: the curve is pretty flat. In other words, there is some correlation, but it’s weak. As Zagorsky describes it, “the average income difference between a person with an IQ score in the normal range (100) and someone in the top 2% of society (130) is currently between $6000 and $18,500 per year,” or roughly on average just $12,000. That really isn’t a lot. In Zagorsky’s words, “the relationship is not very strong.” And though there is a stronger correlation at the highest incomes, few are so lucky, and the correlation is only notable for any IQ above average, after which more IQ makes little observable difference. Thus, “people with above-average IQ scores (> 100) are three times as likely as below-average IQ individuals to have a high (> $105,000) income,” that describes almost no one (only 10% of individuals earn so much), and all one needs to have so good a chance at that is any above-average IQ. (See also studies by D.H. Feldman and Richardson & Norgate; and the economic work of Robert Frank in Success and Luck: Good Fortune and the Myth of Meritocracy.)

It’s thus again just as the PBR models predict: those who end up at the top will be mediocre or slightly above mediocre; not the best and brightest. Observe Zagorsky’s table: look at how many high IQ people earn less than $30,000 a year, which is less than the U.S. national median; indeed, look at how many earn less than $40,000, the national median for those holding a full time job. Indeed almost all high-IQ people earn less than $60,000 a year, which is below the U.S. national median household income. And yet see how many low IQ people earn more than these amounts. Once again, we see luck matters more than IQ. Indeed, we know skills matter more than intelligence (though even what skills you are taught is largely a function of luck, e.g. what social class you get born into, what schools you get sent to, what learning disabilities you you are born with, and so on), but when studied we find even skills are overwhelmed by luck in any correlation with success.

Truth & Consequences

David Roberts writes up a good survey of how much luck actually produces everyone’s fortunes (and misfortunes) in “The Radical Moral Implications of Luck in Human Life” at Vox (17 February 2020). And he’s right. “You didn’t choose your genes or your experiences” growing up (and often, not even after growing up), rather “Both nature and the vast bulk of the nurture that matters happened to you.” As Roberts puts it, “By the time you are an autonomous, responsible moral agent, you have effectively been fired out of a cannon, on a particular trajectory. You wake up, morally speaking, midflight.” And even then you still don’t control all the slings and arrows of fortune to come.

The bad outcomes one might deserve, of course, are those that result from bad decisions you could have avoided. Criminals, for instance, knew better; their outcomes were simply not likely to be a net good, and if they didn’t see that, they do take some blame for that. Even to the extent society shares fault, the individual does too. Even if born and raised lazy, for example, eventually you will always become consciously aware of the cause-effect relationship between that disposition and your fortunes. Unless there is none. Hence, lazy rich people might never notice, because their laziness never really costs them anything, or not so much as they have to care about. Lazy poor people, however, very quickly notice. That’s why it’s actually hard to find an actually lazy poor person. They force themselves to outgrow that disposition, because they have to.

So, one might ask, don’t we need poverty, in order to have that effect? Certainly not. That’s actually in fact the worst of ideas. The negative outcomes of poverty far outweigh any such benefit, because as the PBR model shows, it results in vast amounts of lost and wasted talent, and imprisons or destroys even the industrious. Yes, making everyone rich could be just as bad; but those aren’t the only two options. Once again, the ideal system lies at the equilibrium point: you want everyone to have just enough resources that they won’t be crushed by misfortunes not of their making, but still be lean enough that the ambition to live a life beyond mere subsistence continues to inspire the development of an industrious character. Which is exactly the idea behind Universal Basic Income. It wouldn’t likely encourage laziness, because no one wants to live at subsistence level; being able to only makes it possible to try getting out of that state again, and thus results in more people doing so. Not less. (We now have some evidence to back this expectation up, which I’ll close with.)

Roberts is careful to note that even though almost every advantage we have is a product, ultimately, of luck and not effort, there is still a percentage of it for which our effort matters. And we do need to incentivize people to apply that effort, indeed even become the sort of people who do that. But there are many far more effective and efficient ways to produce that incentive than firebombing entire populations with “misfortune grenades” in the vain hope it will inspire a mere few of them to excel. The collateral damage that inflicts upon the already industrious is alone too great. And this is a negative not just for vast numbers of individuals, but the entire society: all that squandered, unactualized, ground-under, or stalled talent, which is already eager to put in every reasonable effort, could have been contributing to our national productivity. Which benefits everyone. Including—indeed especially—the rich.

Moreover, we need more resource distribution to make even such incentivizing productive. As Roberts correctly puts it:

To become a better person is, at least to some degree, to consciously decide what kind of person one wants to be, what kind of life one wants to lead, and to enforce that meta-decision through day-to-day smaller decisions. They say you are what you do repeatedly; our choices become habit and habit becomes character. So forming a good character, becoming a good person, means repeatedly choosing to do the right thing until it becomes habit.

But “not everyone has equal access to” the “conditions” needed to do any of that. “Neither the capacity nor the need for self-regulation is distributed evenly or fairly,” Roberts points out. Those with privilege and advantage don’t need to put in any work on themselves; and many of those who don’t, can’t because they lack the privileges and advantages necessary to accomplish it. “In a dark irony,” Roberts says, “we demand much more” self-improvement “from those—the poor, the hungry, the homeless or housing-insecure—likely to have the least access to the conditions that make it possible.” As he points out, “Just one more way it’s expensive to be poor” (as has been explained in The Economist, The Atlantic, MoneyWise, Time, The Daily Kos, and beyond). Do you know what does give someone the conditions to self-improve? Access to resources.

It’s thus crucially important to how you see the world, and how you plan to improve it, to admit that luck and opportunity matter a lot more than effort, talent, family culture, or breeding. Indeed, talent, family culture, and breeding are themselves a form of luck; as is even effort, to the extent that it requires skills, and mental and physical health, and an environment that both teaches and encourages their use. And this matters to all of society, and thus to you, even more than it does to the individuals concerned. Making the rich richer does not lift all boats. To the contrary, it sinks more than it lifts. Do you know what does lift all boats, rich and poor alike? Unleashing more talent and productivity already untapped in the population. Only an increased access to resources for everyone can accomplish that.

As Roberts also points out, people are very loathe to admit any of this, and build for themselves elaborate delusions and mythologies to deny the outsized effect of luck on their own and others’ fortunes. Indeed, as Clifton Mark documented, such A Belief in Meritocracy Is Not Only False: It’s Bad for You. And yet much of capitalist and libertarian rhetoric rests on such delusions and mythologies. But if luck determines success to any degree at all, there will always be some percentage of successful people by chance alone—who will interpret their success as due to their unique talent and hard work, when in fact plenty of other people less lucky are as talented or hard working or more. Thus, “being successful” is never evidence you were more talented or hard working than anyone. It often won’t even be evidence you were talented or hard working at all. But even if you are both, it’s being so plus luck that made you successful. It therefore does not follow that the unsuccessful weren’t talented and hard working, rather than merely unlucky. And that is where injustice lies. If there were a God, the injustice would be his. As there is no God, the injustice is ours, insofar as we have any reasonable way to remedy it.

The need to deny reality here drives a lot of ideological nonsense. The Bell Curve argument, for example, that rich white people are breeding themselves into a master race of exceptional talent and leaving the poor and nonwhite populations to flounder under genetically declining cognitive skills, is wholly refuted by even its own data: once you control for all known sources of “luck,” the variance in IQ between black and white populations is a mere 5 points, which is less than the margin of error of any IQ test. And even that trivial variance is likely explained by sources of luck we aren’t controlling for in our analysis (for example, differential exposures to lead in the environment was not specifically controlled for).

Thus the Bell Curve‘s thesis that white people are breeding themselves into a master race is false (for a great deal else wrong with it, see the exceptional analysis on YouTube by Shaun). The differential advantages of white people over black have been in place for hundreds of years—and still have produced no significant difference in IQ. The breeding thesis is thus false. It’s luck that is making the difference—in this case, the mere happenstance of whether you are born white or black in the social system we have created. Skin color is not a choice anyone makes; it is not the product of diligence or talent. It’s objectively random. And there is also nothing born into people of a certain color that decides their abilities and dispositions, either; that is entirely a product of the culture and social system they are born into. Much of which would be mitigated by a bump in resources, providing a lever to overcome those disadvantages with even a little applied effort.

Test Case: Housing

Housing opportunity tests this in practice: numerous studies show “Living in certain neighborhoods seems to expand opportunity, and living in other neighborhoods seems to diminish it,” as noted by Dylon Matthews in “America Has a Housing Segregation Problem” at Vox (4 August 2019). Matthews discusses in particular two studies by Raj Chetty: the Chetty-Hendren-Katz survey “The Effects of Exposure to Better Neighborhoods on Children” in American Economic Review 106.4 (2016) and Raj Chetty et al., “Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States” for the Equality of Opportunity Project (2014).

Kids who grow up in better neighborhoods see their high school dropout rates plummet and college attendance rates double. They are “also likelier to earn high wages and to be employed full time once out of school.” In some cases “the effect size on income was large: about a 31 percent increase in earnings by their mid-20s.” Indeed, the role of luck is so profound, studies find that “opportunity (measured as the share of poor kids who wind up in a higher income bracket as adults) varied widely not just from city to city, but from city block to city block.” Researchers have thus concluded, “Moving to a better neighborhood really [does] make kids better off as adults.”

This is significant because this is not a choice kids can make. Where they grow up is assigned to them by fate, not by any effort or talent or decisions of their own. So that so simple an intervention as relocating poor kids’ families to more prosperous neighborhoods has such a measurable effect on their future success demonstrates how crucially important it is to give people opportunities, rather than expecting everyone to create them on their own. It also illustrates how much untapped talent there is in society that our economic system is not benefiting from, because it is being crushed under by chance accidents—chance accidents society could be preventing or compensating for. Just as we see with housing assistance programs that simply steer needy parents to better neighborhoods—and help them pay the resulting rent, essentially the very thing UBI would do.

Indeed, analysts find that these assistance programs “come pretty close” to paying for themselves in future gains in tax revenue alone—as the benefiting kids grow up to earn more money, and pay more in taxes over the rest of their earning life. Which illustrates the importance of “return on investment”: no matter how many “loser” kids don’t benefit from such help, so many industrious kids do, that we almost make all our money back on our investment. And this is with respect to only a single intervention: simply relocating where kids grow up; and counts only one outcome measure (direct tax revenue from the aided individual, rather than the overall tax revenue gained from the resulting increased productivity and economic activity). Give people the economic power to adapt and to innovate the ways they improve their access to opportunities, and their ability to benefit from them, and to weather more misfortune than they’d otherwise be able, and the overall effect will multiply.

Test Case: UBI Studies

In a sense, UBI, or Universal Basic Income, can be defined not just as cash payments, but everything we all might get equally for free, much of which in fact we already do (more so in other first world countries even than in the U.S.), from police and fire protection, to publicly funded education, to health care and unemployment insurance. All of it helps every one of us weather misfortunes and develop our latent talents, thus leaving more of us able to leverage opportunities into success without being ground under, or stuck floundering. A cash UBI would only supplement most of this, and replace some of it.

Objections to UBI can now be answered empirically. And as we’ll see, only one factual objection remains legitimately debatable: cost. All the others have been refuted in practice and observation. Philosophical objections, of course, relate less to facts and more to values: what we think we should do to organize our governments, societies, and economies as a people. I’ve been addressing that side of the issue throughout this article and will tackle a few more aspects of it below. But objections based on straight fact claims, like that UBI will increase unemployment or make people lazy, evidence has already amply disproved.

The SIME/DIME study completed in the early 1970s is sometimes cited as evidence against the effectiveness of UBI, but it actually isn’t. That study model wasn’t actually a UBI, for one thing. But also, it didn’t actually measure productivity. Mere counts of number of hours worked tells us nothing as to the value of the labor produced; and that overworked people will work less when they can afford to is not a negative result. Nor did it track any other relevant outcome measure, such as impact on crime, or health and happiness, or the net effect of the money distributed on the whole system, i.e. economies grow when people spend money, so the more money people can spend, the greater the impact on economic growth, nationally and locally (a fact too often lost on conservatives, no matter how much liberals keep explaining it to them). And yet those 1970s studies showed little reduction even in what they did measure: participants worked only 10% or so fewer hours, and mainly only because they could spend one or two weeks more at home per year—as opposed to everyone, or any significant number of people, staying home and not working at all, which never happened (ditto).

Those studies were also performed in a completely different social, economic, and political environment. For example, the study’s authors embraced the sexist assumption that gaining the economic freedom to divorce was a negative outcome; and the participants largely assumed a sexist division of labor in families was to be preferred. The authors also assumed having the economic freedom to leave an undesirable job and hunt for one more tolerable was a negative outcome, when in fact labor mobility is indicative of a beneficial dynamic for any economy (a point I’ll expand on in a moment). That study also took place in a period with significantly lower income disparity, and it is the latter that increases the impact and importance of UBI. (There were other problems with the SIME/DIME study, discussed by Marinescu, cited below; and as also summarized, again, by Dylan Matthews at Vox.)

By contrast, Sigal Samuel surveys numerous experimental studies of more recent vintage and more informative design in “Everywhere Basic Income Has Been Tried, in One Map” at Vox (19 February 2020), including little known examples of real UBI (in Alaska and Native American tribes, for example; and we just ran another one across the whole United States in 2020-2021 that got the same results; and this plethora of passed tests is a point of discussion now). See also Karl Widerquist’s website and book; the research summary of experts Joseph Hanlon, Armando Barrientos, and David Hulme in Just Give Money to the Poor; and the ongoing GiveDirectly crowdfunded research. The results are fairly consistent across dozens of experiments (and wholly refute contrary reports, which, when they even make any relevant arguments at all, ignore the actual evidence or make assertions wholly without evidence):

• “The evidence so far suggests that getting a basic income tends to boost happiness, health, school attendance, and trust in social institutions, while reducing crime.” It also improves other things, like high school graduation rates, homelessness and even employment; and where tests have been large enough, they’ve even improved the economic status of entire communities not even part of the study through multiplier effects.
• “The programs analyzed suggest either no effect on labor market supply, or a slight reduction in work and earnings. The evidence does not suggest an average worker will drop out of the labor force when provided with unconditional cash, even when the transfer is large,” in fact even “recipients of windfall cash disbursed annually,” i.e lottery winners, “do not drop out of the labor force” (quoting Ioana Marinescu from a cited study from 2017 “No Strings Attached: The Behavioral Effects of U.S. Unconditional Cash Transfer Program”)
• Every study measuring them found substantial improvements in the health and nutrition of recipients, including children, illustrating that any UBI will actually produce a savings offset in healthcare.
• And in every case so far observed, UBI works as well or better than all the existing welfare systems it would replace.

Often not considered are the hidden positive effects of a real UBI. Currently in the U.S. we are trying to “privatize” the intentions but not the reality of a UBI with minimum wage laws. But those laws would be rescinded under a UBI regime, since UBI would be replacing their function. The effect would be a considerable benefit to employers. Large businesses would enjoy slightly reduced labor costs (this is also a principal effect of a national healthcare system, as discovered by every nation that has implemented one), while small businesses and entrepreneurial startups would enjoy large reductions in initial labor costs. For example, you could contract labor for pennies on the dollar in exchange for perks, discounts, benefits, or profit shares, or even contract labor for free in exchange for promises of future payouts upon the hitting of revenue targets. Which flexibility could expand small business growth and innovation, without contributing to poverty or wage slavery. It would also create jobs.

In fact, because UBI eliminates wage slavery, it eliminates all the concomitant evils of wage slavery as well: businesses now can try to shirk on workplace safety and tolerability, and treat employees like crap, because no one can afford to lose even a shitty job. But UBI will provide workers the economic freedom to not be trapped in such arrangements. The net effect will be: such arrangements will disappear. As employers can no longer retain a workforce they mistreat, workplace safety and quality will substantially improve. This will have further net effects on the health and education outcomes of workers, further reducing costs elsewhere in the economy. Of course, Andrew Yang has also made the point that UBI is eventually going to be inevitable as automation displaces human labor, eliminating the traditional mechanism of generating household income.

It’s also important to remind anyone doubting the merits of UBI, that virtually every dollar of UBI goes directly back into the economy. Whereas rich people sink large quantities of cash into “money sinks” (from savings accounts to hedge funds and other forms of stock speculation, or trading in big ticket items that have of themselves a relatively smaller impact on an economy, e.g. trading priceless works of art, buying yachts; and, of course, parking money in offshore tax havens), ordinary people spend their income on goods and services en masse. If only rich people used Amazon or Target or McDonalds or even your average corner nail salon, those businesses would collapse. In fact, nearly all businesses would collapse. And with them the whole economy. No more rich people. But if an entire population started spending, say, 10% more money on goods and services, every business increases its economic activity, increasing gross domestic product. Since nearly every dollar of UBI goes back into the economy, particularly local economies, it actually isn’t wasted, but has enormous collective benefit. It benefits communities, not just individuals.

This would be most obvious in the countless dying rural towns all over America. Think of all the money that would now go into those economies, as locals start spending their UBI, creating jobs and significantly increasing economic activity and growth. It’s almost the single most effective thing you could do to rescue small town and rural America. But the same effects will be observable even in urban centers, particularly poor neighborhoods. But it is in rural America the effect will be most felt. Because one other effect of UBI is that, being universal, its recipients will benefit from moving to communities with lower costs of living. The brain and labor drain on small towns and rural states would reverse—growing their economies, and thus general prosperity, even further. A similar benefit of UBI is the advantages it provides us in the event of national emergency and disaster—such as our present pandemic response, but also regional equivalents such as hurricanes and earthquakes. Having UBI already in place would make the United States and its population and economy more resilient, not less.

Finally, too, as Yang’s analysis shows, UBI would replace much of the piecemeal welfare system we already have, offsetting its net cost. Not only dollar for dollar, but also with every dollar moved, the administrative costs of those other programs will be eliminated. And as those tend to be conditions-based, their overhead (in vetting and auditing) is much higher than a simple UBI program’s would be. The cost of even basic UBI is nevertheless quite high. And I think this is the only real criticism of it that holds up: like all the useful luxuries of civilization, a thing you should have, you only should buy when you can afford it. We should as a community fund fire fighting, for example; but only if we as a community generate enough wealth that we can safely afford it. And so on down the line of every wise move civilizations have made (from universal education to, as in all other first world nations, universal healthcare). So we should only implement UBI, obviously, when we can actually afford to. So the question becomes: Are we there yet?

Paying for It

Andrew Yang of course developed a simple guide to the concept of UBI as part of his Presidential campaign, which covers most questions one might have about it—including how we’d pay for it, although his plans in that respect are unlikely to work as he claims. He is right that the 3 trillion dollar cost of his UBI proposal (of basically sending $1000 per month to every adult citizen) would replace existing expenditures, and in fact more than he calculated. Updating Yang’s numbers, UBI would replace roughly $800 billion in other programs he counted (from welfare to unemployment insurance), but it would also replace about $800 billion in social security expenditures (since social security payouts would not add to UBI, but only make up any difference in average monthly benefits, which are already above $1000) so UBI’s net cost in the U.S. would be “only” 1.4 trillion dollars. But that’s without a national healthcare system, which we also ought to have, and also has to be paid for. That would cost roughly another 2 trillion dollars (after offsets and such are tabulated, e.g. such a system would replace medicare and medicaid). So actually, we’re looking at $3.4 trillion a year in new spending, for a standard social safety net every other first world nation already has—and UBI. (I had previously written 2.4; the following has been revised to address the correct amount.)

If you calculate from IRS data, the entire collective incomes of the top 1% of earners (which means roughly everyone who earns more than half a million dollars a year) is just over 2 trillion dollars. If we surtaxed all income above half a million dollars at a flat rate of 50% (which means in addition to existing income taxes subject to deductibles and so on)—and I think that would be entirely reasonable—we’d bring in new revenue of about 1 trillion dollars. And a national sales tax of 15% (called a VAT, which would be pre-built into advertised prices rather than added at the register) could raise about 1.73 trillion a year. Many other successful nations have just such a tax, so we know its effects on economies are not prohibitive.

So those two revenue streams alone would make up all but 670 billion of the dollars needed. We already know improved enforcement of existing tax laws would bring in hundreds of billions a year—indeed, $500 billion according to one recent study (and another). So obviously we should be doing that. Which leaves only 170 billion to account for. So the question then becomes, is it reasonable to gain the corresponding national benefits with a 60% “absurd income” tax instead of only 50%, adding another 200 billion dollars to our national revenue? I believe so.

The already existing budget shortfalls of almost a trillion dollars a year would gradually be made up if we returned to a pre-Reagan income tax regime (canceling all Republican tax cuts then and since would raise over $380 billion a year in current dollars), ended the war on drugs (for at least a $70 billion net gain), and substantially cut our spending on useless foreign wars (to the average tune of $320 billion a year) and corporate welfare (by the narrower definition, in adjusted dollars, gaining us some $70 billion), and enacted a reasonable drawdown in overall military spending (earning back at least $117 billion).

UBI will also replace roughly a hundred billion dollars in federal employee costs by simply not duplicating UBI to federal employees and pensioners, i.e. if a federal retiree is receiving a pension of $1500 a month, or a federal worker is receiving a salary of $1500 a month, they would continue receiving that instead of UBI. Statutorily, UBI would simply be a part of their already-promised compensation package. Similarly, all the federal government’s expenses on health insurance for its employees would be replaced by a national health care system, thus generating an additional savings offset. With over 3 million employees, civilian and military, that’s 36 billion dollars in salary replaced by UBI; and current federal health care benefits total about $90 billion for civilian employees (active military already have free healthcare). So the total offset will be 136 billion dollars.

All of that erases $893 billion of our current $966 billion deficit; but $393 billion of that deficit comes from paying interest on previous accumulated debt, a quarter of which (almost $100 billion) is actually interest the government owes to itself. We could actually suspend interest payments on that debt (essentially “renegotiate” that debt to zero or near-zero interest; and when we get to an economy good enough to start earning budget surpluses again, the overage would go directly back to those programs the principal was borrowed from until it’s paid down—which means, pretty much, social security, federal retirement, and medicare, the very programs the UBI-plus-healthcare system would already be subsidizing). That gets us to $993 billion, more than we need. A reasonable means-based copay on prescriptions and non-emergency medical services would earn even more.

With all that—a return to our highest-earning tax regime and lowest military budget since Reagan took office, and an end to draconian drug laws and corporate welfare programs (at least any that aren’t revenue neutral or actually earn a profit), renegotiating the debt the government owes to itself and enacting a graduated copay system in the healthcare system we are otherwise subsidizing, actually enforcing existing tax laws, and then accounting for all the offsets (all the things our new system will already replace)—we could actually balance the national budget as well as have national healthcare and UBI. And all we’d need add besides is an “absurd income” surtax of 60% and a 15% VAT.

Maybe that’s too much. But really, it’s hard to claim so. Such a VAT is already standard in much of the first world (indeed a 15% rate is common); and the difference between someone’s million dollar annual after-surtax income dropping from $750,000 (at a 50% surtax rate) to $700,000 (at a 60% rate) is hardly anything to honestly complain about; and incomes in the millions or billions, even less so: at the end of the day, you still take home millions of dollars a year.

I normally am against sales taxes for their disproportional cost to the poor, but that impact is mitigated precisely by UBI. So there is no remaining objection to enacting the same kind of VAT most other first world countries already have. And there is not a really good case to make that someone who earns a million dollars in a year (who is thereby essentially hoarding resources), should not return to the public good a quarter of it more than they already would (under our revised pre-Reagan tax regime), considering all the benefits that accrue from UBI even to said millionaire. And yes, it would just be a quarter of it more; a 50% surtax on surplus income above half a million dollars is a quarter million dollars. Yes, mega-millionaires would be giving even more, but once you get above accumulating a million dollars a year, it’s no longer credible that they “need” all that surplus. Even the argument that millionaires need it to invest, because, being millionaires, they are “obviously” smarter investors or something, we just saw is based on a false premise: there is no evidence millionaires are so much better at investing as to outperform the effects of investing instead in a national UBI and healthcare system—which, again, they benefit from: in reduced labor costs for all their businesses; in a happier, healthier, and more secure (and thus more productive) workforce and citizenry; and in all the consumer spending UBI generates; among much else, such as reduced crime, a more skilled workforce, and so on. All as I just enumerated.

So we could do this.

There are of course other ways to pay for a UBI-plus-Healthcare upgrade to our social system, and real economists would have to work out which was actually best; mine is simply what seems most feasible to me, and for which we have the most evidence regarding impact.

There are also other things we could implement along with UBI that would answer those who believe everything ought to be earned no matter what the consequences of that ideology on society actually are. For example, we could tie UBI to national service. We would exempt the disabled, of course—those with physical or mental health conditions that render them unable to contribute to national service—on the same principal of disability insurance we already implement, wherein all our tax contributions essentially “buys” insurance against just such a fate for ourselves and our dependents. We would also expand “national service” beyond the armed forces, so even conscientious objectors could contribute.

Imagine, for instance, we split the whole of our military between hospital corps and combat corps, such that the hospital corps serves all forces and would be by law unarmed (combat corps would provide their physical security), so even pacifists could serve, and even help us fight wars by helping care for the injured and sick. We already have an ambassador corps, peace corps, americorps, and a lot else, plus our entire diplomatic foreign service, and any government service whatever—which could all count as “national service” for purposes of earning a lifetime pension in the form of UBI. We could also expand programs like americorps, and recognize state and municipal initiatives of like kind, implementing community service jobs that range from cleaning up neighborhoods to assistive data entry or phone banking from home, to other administrative, custodial, or more specialized jobs of all kinds.

In such a system everyone upon reaching the age of 18 would begin receiving UBI until the age of 38, after which year they would have to pay all that UBI back over the remainder of their life through a special income surtax, unless (or until) they’ve completed any accumulation of 2000 hours of national service. That way, veterans of course would always earn lifetime UBI; but so would anyone who serves in a noncombat capacity in jobs that serve the public good. Moreover, anyone in noncombat service completing national service hours to earn their UBI could be required to do so at reduced pay (the same way our military already does: private combat forces always earn significantly more). It would be no hardship, as they would already be receiving UBI, after all; and to earn that for life they could work at an even lower salary than those who continue to serve after completing their 2000 hours. Many of these national service jobs (like local community service work) would even be volunteer, or minimal pay (beyond the UBI already being received). The net effect would be an even more affordable government.

Conclusion

All these findings demonstrated in practice and through systems modeling corroborate counterintuitive results: since luck is more important than talent in a completely free market system, completely free market systems actually tend more toward rewarding mediocrity and punishing talent and effort than enthusiasts believe, and justify wealth hoarding by the few even less than most had already thought. All this evidence has changed my mind on a few things. I was uncertain about the effectiveness and justification of universal basic income as a societal strategy, but I now think UBI-plus-capitalism is actually the best social system to implement. And as I ever have, I derive this conclusion from the evidence rather than ideology.