Social Welfare, Total Welfare

Lately in my comparative politics class we have been discussing measures of social welfare like the UN Human Development Index. The idea is that maybe our concept of "the good" should include not merely GDP per capita but also factors like social inequalities, survival during disasters, poverty, and government social spending and policies. The HDI in particular includes log GDP per capita, education as measured by literacy and enrollment in primary, secondary and tertiary education, and life expectancy at birth as predicted by current mortality levels.

Amartya Sen summarizes, "Human development, as an approach, is concerned with what I take to be the basic development idea: namely, advancing the richness of human life, rather than the richness of the economy in which human beings live, which is only a part of it."

Fair enough. However, I've got a different criticism. Why do we assume that total population is not important? If we believe that human lives are good, then we should conclude that more human lives are better. But normally in the economic literature one hears about fertility or population growth having a negative impact on GDP per capita, as if the latter was obviously the primary goal.

Jeremy Bentham's saying that we should do the greatest good for the greatest number comes to mind--along with its perennial weakness in not specifying how to calculate that.

I think that an improvement over maximizing GDP per capita would be maximizing the total GDP. This is the product of GDP per capita with population and thus takes both sides of Bentham's phrase into account. (Other and better utility measures surely exist. But let's proceed.) What advice would one give countries which sought to maximize total GDP? Considering the entire world, global GDP rose over 3700% in the 20th century and GDP per capita over 860%. Thus there were substantial components to global GDP growth coming from both improved living standards and increased population, with living standards contributing most.

Looking at the current PPP-adjusted leaders in total GDP indicates that it encompasses a wide variety of combinations of GDP per capita and population.

One obvious question if we maximize total GDP is whether Malthusian outcomes ought to be ruled out. Personally I am not convinced that having a huge but poor population is a priori worse than a small and rich one. However, since more of global GDP growth in the 20th century came from rising living standards than population growth, simply expanding population seems a doubtful route to large GDP. Also it seems difficult to imagine any government actually endorsing this given the sacrifices it would mean for people alive today.

But that still leaves many other options open. The more interesting question is whether you clearly prioritize increasing living standards over population growth or whether you give comparable weights to both. Global catastrophic risk also enters the picture here, since depending on which risks you fear, you may believe that one or the other of these options would be less dangerous.

Making Up Words

I'm going to use the term "pro-human" to describe a stance on transhumanism similar to Nicholas Agar's. The definition of a pro-human position is that:

We should plan a future with a growing number of high-quality human lives. The creation of posthumans is likely to conflict with this and should therefore be avoided.

(By "human" I mean a member of Homo sapiens and not the expanded definition that many transhumanists use.)

The closest term now in existence seems to be "bioconservative." And that term describes my views too--conservatism is about a slower pace of change, and I think many so-called enhancements would change the world for the worse.

Pro-human has a different emphasis than bioconservative, however, one on the value of our shared humanity as opposed to the past. It is more general in including people who think that "moderate" enhancements inside the bounds of current human limitations might be OK. It may also be more accessible for people new to this debate, who may assume that bioconservatives are conservative in its more everyday political sense, like neoconservatives. In fact, some bioconservatives are fierce political progressives. And finally, the term pro-human immediately makes the point that humanity is a biological species and that enhancement could remove us from it. That should be obvious, but Ray Kurzweil has disputed it.

So pro-human highlights potential catastrophe that remains farther off, whereas bioconservatism also includes nearer-term changes that would not threaten humanity by themselves.

Lights in the Tunnel

I just finished the first half of Martin Ford's book The Lights in the Tunnel. It's good reading and I plan on buying a copy at some point. In fact, it's been distracting me from studying for my econometrics exam, so I probably should stop soon.

I couldn't help but feel that the major premise of his scenario had a hole somewhere. He predicts that increasing automation and robotics in the economy could lead to rising unemployment, falling consumer demand and consequently a massive permanent depression. Robin Hanson points out what seems to me the flaw here: demand doesn't have to come from ordinary consumers. In Hanson's words:

Ford’s mass-market theory of production is nothing like standard economic theory.  Sure high income inequality might be ethically bad, and threaten political instability, but it does not at all threaten economic collapse – producers can focus on giving the rich what they want, and innovation and growth is just as feasible for elite products as for mass products.

Now, inequality is a problem. But it's potentially a solvable one; you can work it out through political channels, even if success is by no means guaranteed. The future problem that seems to me fundamentally unsolvable is the advance of human biotechnology and intelligence enhancement. How do you work out material inequality, or much anything else, when the enhanced are on an entirely different mental plane from ordinary humans? I don't think you can. 

Those issues seem absent from Ford's book. Perhaps it's no coincidence then that he offers typically unqualified support for technological change.

Transhumanists vs. Feasibility Skeptics

Will Oremus posted a response in Slate to Geoffrey Miller's recent interview. Several experts, including Lee Silver, apparently state that we are not near being able to genetically select for IQ because thousands of alleles influence it. These people's credentials are pretty impressive but their argument seems questionable, at least from the detail presented by Oremus. Yes, it's clear that IQ is coded through numerous alleles. But identifying a large fraction of those alleles is exactly the BGI goal!

That's the short of it. The BGI project is also more clear to me from watching this video. (I actually watched the entire 90 minutes. That must be my record for Youtube.)

I think the sensible viewpoint is actually missing from both sides of this debate. On one side are transhumanists who agree that there is potential for genetic IQ pumping within a decade, like Steve Hsu and Geoffrey Miller, and actually favor doing that. On the other side are people like Lee Silver and Razib Khan who are skeptical it can work. What we need is more people like Bill McKibben who appreciate the potential of these technologies but understand how profoundly dangerous they would be.

We need to start weighing whether research is actually worth doing if it could send us into the kind of extreme arms race McKibben described. My opinion is that the risks of exploring the genetic basis for IQ clearly outweigh any potential benefits. This BGI project should be shut down.

Remember the words of Albert Einstein. In 1932 he said, "There is not the slightest indication that nuclear energy will ever be obtainable. It would mean that the atom would have to be shattered at will." Experts can be wrong. And if they are, we might soon find ourselves staring at a new Manhattan Project. 

Best Time to Be Alive?

Many people have said that this is the best time to be alive in history. And on the surface, it seems obviously true. World GDP per capita rose over eight times in the last century. Who would not want to share that? Life expectancy has shot up too.

But I wonder if this is basically just luck. In other words, John F. Kennedy thought that the chances of war with the Soviet Union were between 1 in 3 and even during the Cuban Missile Crisis. A bad outcome did not materialize but if it had then the world would look very different. 

It may not be that our expected quality of life is increasing as much as we think. In fact, I think it may have peaked at some point in the past--a kind of golden age. Outcomes for different nations and individuals may now be linked together much more than previously, and collectively we may be happily experiencing the more prosperous end of what is possible. But technology may have the effect of increasing the variance or the spread between the best and worst possible outcomes. If this is true then there is nothing to prevent the world as a whole from being unlucky in the next round of poker. 

IQ Economy?

I came across this story in Forbes a while ago by Rich Karlgaard, "The Trillion Dollar IQ Business." It's all very fascinating, this idea that high IQ people are inheriting the Earth...but also wrong, at least for now.

Economics Research Topics

There is still some chance I will do an economics PhD, so I thought this 80,000 Hours page about high-impact economics research questions was quite interesting. The major topic on the table seems to be economic growth. If you read an economic growth textbook by Robert Barro, he would agree.

However, while most people and especially economists would probably find it bizarre, I suggest that economic growth may not be good for the world right now. The problem is that it contributes to risky technological change. The real priority should be minimizing existential risk. To me there is therefore an extremely acute need in the economics profession to undertake meta-research discovering which questions most address this.