Paul Graham’s Second Inequality Essay is Only Slightly Better. Here’s What’s Still Wrong With it.

I enjoyed Paul Graham’s revised, reduced essay on income inequality. It is, in many ways, better. He invites us to compare it with the first essay, and I did. It’s quite welcome that he removed the ridiculous straw man argument that so upset me the first time around. It also removed the unfortunate references to being hunted. Jason Calicanis said it well, according to Sarah Lacy: “If you’re lucky enough to be rich, shut up and enjoy it.” One of my personal golden rules is somewhat similar: “If you’re arguing a point of view that helps you personally, there’s a very good chance you’re wrong.” So, thanks to Graham for cutting all that out. It wasn’t helping the core argument. For the purposes of this essay, we’ll only look at the revised shorter version of Graham’s original longer essay.

All that being said, Graham does not seem to understand that the phrase “income inequality” is, essentially, a unit of measure. It is not a state of being. It’s not binary. It’s a level. It goes up and down. You can no more attack income inequality than you can attack the temperature. What you can attack is high levels of it, dangerous levels of it.

Perhaps a good analog is global warming. Yes, we can attack global warming, as a turn of the phrase indicating a planet that is getting too hot. What we cannot attack is the temperature. We cannot attack warmth. The distinction is subtle, but important. Graham is conflating the two. And he is using this conflation to mask some fairly fallacious and odd arguments.

So then, perhaps we need a slight re-editing of his essay, in which the unit of measure is switched to another one, will help illustrate the problems with his argument.

Lots of people talk about global warming. Nearly all say it is bad if warmth increases, and that it would be better if it decreased.

But warmth per se is not bad. It has multiple causes. Many are bad, but some are good.

For example, factory pollution rates and oil well fires are bad things that increase warmth.

But space heaters also increase the warmth. A manufacturer whose space heater business succeeds will end up with a lot of warmth.*

And unlike high factory pollution rates and oil well fires, space heaters are on the whole good.

Since warmth per se is not bad, we should not attack it. Instead we should attack the bad things that cause it.

For example, instead of attacking global warming, we should attack dirty air.**

Attacking warmth would be doubly mistaken. It would harm the good as well as the bad causes. But even worse, it would be an ineffective way to attack the bad causes.***

We will not do a good job of fixing the bad causes of global warming unless we attack them directly.****

But if we fix all the bad causes of global warming, we will still have increasing levels of it, due to the increasing power of technology.*****

On it’s own, with changed adjectives and nouns, the silliness of the essay becomes readily apparent. Some of it is tautological and some of it is nonsensical. But dig a little deeper, and it becomes even more weird.

Some important notes:

*This is where the essay really takes a subtle turn, right? We could leave the “lot of money” phrase in, because it first blush it seems like it’s not really part of the “warmth” issue. But of course it is, because he’s tying individual wealth/warmth with universal wealth/warmth. By changing the adjective, one starts to see how silly this is.

**It could be argued that right here is where the argument goes off the rails. Graham could have used one of the two “bad things” he already mentioned. For the sentence right before this says “Instead we should attack the bad things that cause it.” He listed two bad things: high incarceration rates and tax loopholes. In his example, then, it would have made sense to use one of those! But he didn’t. He chose poverty. This is key. He chose poverty in an endeavor to muddle his argument. Because poverty is not actually a bad thing that causes income inequality, it is a circumstance of it. He is conflating the two. Which is why my example of pollution stands. It sounds good, but really, what causes it? The other bad things that we want to attack. In my case, polluting factories, in his case, tax loopholes (i.e. government tax receipts) and high incarceration rates. Great! Yes! Let’s fix those.

***This is a good point, though not an especially controversial one. What’s really going on, underneath this sentence, is the proposition that a certain root cause should be categorized as good not bad. Fair enough. To use my global warming example, we’d probably not unduly spend all of our time focusing on space heaters. We would however, perhaps pass a couple regulations to ensure that their manufacture was more responsible, and they weren’t, say, needlessly polluting the atmosphere. But hey. Not the main problem. What Graham is really saying here is that he is advocating for a certain cause — “founders” to be classified as a good cause and not a bad cause. That is, call founders more akin to space heaters and not factory polluters. This is completely reasonable. In tackling the problem there would have to be a whole big sorting of good causes and bad causes. People will argue points. I’d love to read a Graham essay explaining why founders who hire lots of people should be exempt from high taxes on high levels of income. I’m not against it.

**** Amen. I don’t think anyone disagrees with this. I would put on the “bad causes” side of income inequality low tax rates on high levels of wealth, carried interest loophole (on venture investment as well as trading), a low inheritance tax, an inefficient system for delivering benefits to the poor, increasingly non-meritocratic schools and the ability for the rich to easily jurisdiction shop around the globe for favorable tax rates. Boom. Great. Yes. Lets fix those. This is the whole point. An anonymous YC founder friend of mine demolished Graham’s essay in a much shorter manner by touching on this point: “His argument is “Everyone thinks that the room is too cold and they blame it on the open door. They don’t blame it on draft coming through the door. Let’s fix the draft to make it less cold in this room.”

***** This is as absurd in my global warming example as it is in his startup example. A startup is enabled by technology, yes, but at it’s core it’s just a business. The technology aspect on its own has nothing to do with whether or not it contributes to the problem of income inequality. Yes, technology is powerful, changes the world and is disruptive. But does technology, on its own, make income inequality or global warming better or worse? It doesn’t. We control technology, we put rules around it, we can harness it for good or evil. Its scientific progress is inexorable, but its existence is empowering. It’s a weird thing for a VC to say “well, technology exists so we can’t do anything to fix something!” Isn’t that the entire point of Silicon Valley? To harness technology to solve the world’s problems? How weird is it that a VC is using technology as an excuse to not be able to do something. And how depressing.

Conclusion

My strong desire is for Graham to just come out with it and write the essay he’s really thinking: “if the wind is blowing on these tax raises and we have to start deciding what are good causes and bad causes of income inequality, I advocate to put all VC returns into the good causes bucket so we can keep our lower capital gains rate. Not just good old fashioned founder/early employee equity returns and angel investment returns for individuals, but the fees and carry that venture firms make their living off of, and (maybe? I don’t know how he feels on this one) the returns the LPs make as well. Because all of this money creates jobs.”

It’s not a completely crazy argument, but he seems scared to make it outright. I’d love to hear it, debate it, talk about it. This stuff is important to me. I re-invested my money into tech precisely because it creates jobs.

Those founders, I would agree, should be viewed as a positive economic force. That is very different than the tax rate I pay on my profit as an angel investor, and more to the point it is completely different than the taxes on the return on the carry and fees a VC would pay. I suspect there’s a bit of willful obfuscation here. Because in a lot of ways, a VC isn’t that different from a trader, who Graham has already placed in the “bad causes” file.

So let’s have it out. Is a VC a job creator? Is a hedge fund like Bain a job creator? What about a high-frequency trader? After all, they’re investing in stocks, stocks are equity in companies, companies make jobs sometimes. And sometimes not. And sometimes startups lay people off, or disrupt jobs in other industries. Is Uber a job creator? I have no idea. Let’s debate it all! Some people already are! There are many good points to be made in here: a venture fund is probably gonna create more jobs than, say, a rich person keeping their money in Treasury Bills. Though honestly, I’d have to see some empirical research, hear some different points of view, because, you know, I haven’t researched it much yet! I’d love to hear Paul Graham’s point of view on that topic.