Review of "Doing Good Better"

Forty years ago, and more, I was intrigued to note that a surprising amount of the Law given to the Hebrews in the first five books of the Old Testament is actually very sound business advice.  As only one example, every twenty years business seems to rediscover the idea that it is good business to pay attention to your people.  In the 1960s it was "Theory X and Theory Y."  In 1980 it was "In Search of Excellence."  Around 2000 it was "The Human Equation."  (We seem to be overdue for this rediscovery, at the moment, but that may be due to disruptions because of the pandemic.  Why not? The pandemic is being blamed for supply chain issues and all kinds of other business impedances.)

At the time, one of the things that I noted was that God did not seem to be terribly interested in efficiency.  There are a number of passages where those following the Law are commanded not to be particularly efficient.  I wondered why this was so.  And, forty years later, along came the pandemic and made everything glaringly obvious.  Efficient systems are brittle.  Capitalism's relentless, and single focussed, pursuit of efficiency at every and all other costs, produces systems that are not, and cannot be, resilient.  If something goes wrong, the whole thing comes crashing down.  Efficiency leads to calamity.

William MacAskill wants us to be more effective in our altruism.  He wants to make our charity more efficient.

The book opens with a story about how having a good idea is not enough.  True enough, as far as it goes.  But then we get into chapter one, which is, oddly, outside of the rest of the book, and a section which dances all around, but never really states outright, the idea that we should be as rich as we can be, so that we can donate as much as possible to charity.  (Later in the book the idea of "earning to give" is promoted more explicitly, and repeatedly.)  It is difficult to see where the efficiency comes in that proposition.

MacAskill then gets into the five key questions of Effective Altruism.  Question one is, how many people benefit and by how much?  Question two is, is this the most effective thing you can do?  Question three is, is this area neglected?  Question four is, what would have happened otherwise?  Question five is, what are the chances of success and how good would success be?

Question one, how many people benefit, and by how much, is really just a restatement of utilitarianism.  And it has the problem of utilitarianism: how good is the greatest good and how do you measure good versus number?  MacAskill proposes that we use the quality adjusted life year or QALY.  But this measure relies on an objective scale of how much better your life would be in a certain situation as opposed to another, and these kinds of assessments are generally pretty subjective.  As we say in management (not all together accurately or truthfully) you can't manage what you can't measure.  The measures being proposed here are highly questionable.  MacAskill is saying that we can measure, and we can manage, altruism, but his measures are extremely suspect.

In the first chapters, MacAskill leaves the impression that a few simple heuristics are all that are needed to choose praiseworthy activities, or charities to which to donate.  However, as is shown in chapter seven, actually making these decisions requires enormous amounts of data, not always available to the charity consumer, and requiring a lot of work to analyze and compare for different charities.

In chapter eight MacAskill talks about sweatshop boycotts, fair trade coffee, and other fair trade products, carbon offsets, and other forms of moral licensing.  He cites research that indicates that people who have previously chosen "green" products, have seemingly given themselves a moral license to then lie and cheat on a separate task. This is extremely interesting, but also confusing, research and results, given other studies which have found that briefly discussing moral issues, or even simply asking people to give their word that they will not cheat or will do their best, results in significantly increased moral behavior. Obviously additional research needs to be done in this area.

Chapter nine examines choice of career.  Unsurprisingly, the option of earning to give pops up rather frequently.  However, MacAskill starts out with a major concern that people not follow their passion: he considers this really terrible career advice.  Agreed, following your passion is terrible career advice if that is the only thing you do.  For one thing, your passion can be addressed in a variety of ways.  Personally, I always wanted to be a doctor when I was younger.  I have had a very satisfying career in information security, which, initially, would seem to be a completely unrelated field.  However, a number of the factors and characteristics that a career counselor once pointed out to me did support my desire to become a doctor, also apply to the field of information security.  I'm dealing with a technical field, and needed technical background to deal with it.  I am, primarily, helping people.  I frequently tell people that I do almost exactly the same thing as a doctor: you go to the doctor with a problem, the doctor suggests a solution, you go away and try the solution, you come back and report to the doctor whether it worked or not, if it didn't work the doctor suggests another solution.  I do pretty much the same thing.  (And very often, as with the doctor, when the solution actually fixes the problem, neither of us is entirely sure what the problem originally was.)

One of the areas that MacAskill addresses specifically, a couple of times, is that of entrepreneurship.  He suggests that entrepreneurs should, rather than simply trying something and seeing if it works, do a lot of analysis before they get into a specific product or service.  Unfortunately, that's not how entrepreneurship works.  Yes, there are the entrepreneurs that create a one-hit wonder, and become rich and famous.  But most entrepreneurs start many many small businesses, and for every twenty that they try, nineteen fail.  The twentieth, hopefully, does succeed and create enough money for you to keep going and try another nineteen failures, hoping for the twentieth success.  What with creating all these businesses, there isn't an awful lot of time available for doing analysis, in advance, of all the things that you need to try and start up.

MacAskill's advice on deciding on a career turns out to be remarkably similar to anyone else's advice on deciding on a career: learn everything you possibly can about the career.  It's a very complicated task.  It takes a lot of time.  And a lot of energy.  And a lot of information.  And a lot of analysis.  It's not easy getting it right.  And MacAskill doesn't particularly make it any easier.

(I have to mention MacAskill's comment on Wikipedia.  He says that this is an example of doing good volunteer work: that you can contribute high quality work to Wikipedia, and therefore you can provide a significant benefit to many people at almost no cost to others.  As I have found, to my regret, contributions to Wikipedia are almost pointless.  If you are an expert, you will find your contributions "corrected" by those less well-informed than yourself.  What you will have contributed, is then either reverted to a previous version, or corrected so that it is now error-ridden.  Wikipedia is, as I have stated elsewhere, a form of entertainment at best, and a really terrible source if you are looking for information to make significant decisions.)

The implications of doing what MacAskill proposes, and making altruism effective, are even more bizarre.  If you could, accurately and reliably, measure and calculate the good that you would be doing by addressing a particular charitable endeavor, it would create a rush on the most effective charities.  These charities would be flooded with money, and possibly would solve a number of problems.  However, of course, the other problems would still exist, and the charities endeavoring to address them would be starved for money.  This idea is somewhat addressed in chapter four, which looks at the question, is this area being neglected?  However, the discussion itself is rather simplistic, and fails to address the ongoing recalculations that would have to take place as effective altruism grows in popularity, and more people would be doing precisely the same calculations.  The final outcome of such an adoption of effective altruism would tax the capabilities of even a quantum computer.

The principles of doing good better, and Effective Altruism, do sound good in theory.  They boil down to taking some basic tools of market analysis, and applying them to charity.  However, in actual practice, it turns out that applying these tools to charity is as difficult as it is applying them to an actual business. In addition, the implications of the adoption of Effective Altruism, while it could prevent some of the more egregious stories of charities gone amok, could also lead to significant negative consequences, ranging from competition, at a very intense level, in the charity marketplace, to a kind of paralysis by analysis which would effectively delay donations to charities, possibly by a significant time period.  It is quite possible that the Effective Altruism movement might become one of those egregious charity horror stories.  (Yes, I *am* thinking, in some regard, of the collapse of FTX, and the loss of one of Effective Altruism's big success stories and most attractive champions.)