Let's say you're a relatively well-off American with an income of $100k and have taken the Giving What We Can pledge to donate 10% of your money to the most effective charities you can find. The standard deduction in the US is currently $12k, which means if you were to donate $10k/year it effectively wouldn't be tax deductible. [1]

The standard EA suggestion here is Donation Bunching: instead of donating $10k every year donate $20k half the years and $0 the other half. In the years where you donate you itemize your deductions, in the others you take the standard deduction. Considered over two years, $8k of the $20k (40%) is deductible.

But we could do better than that! You can deduct up to 60% of your income if you're donating cash, so you could have five years of donating $0 and one year of donating $60k. Considered over six years, $48k of the $60k (80%) is deductible. [2]

This sounds great, right? Money donated to effective charities does much more good than money collected by the government, and with a bit of planning you can effectively move 80% of your donations to being pre-tax. But I don't think it's a good idea!

Over time people change, and a very common way people change as they get older is to turn inward. You start out as a bright-eyed idealist, enthusiastic about making the world a better place and willing to make sacrifices for what you believe in. Then you burn out working too many hours doing something that doesn't feel as effective as you thought it would be, or you start feeling a pull to have kids and focus your efforts there, or you just stop feeling so motivated by altruism, or dozens of other things, and after a few years the idea of giving your money to people who need it more still sounds nice but isn't a priority anymore.

While we don't have good data on the rate at which this happens, in a small sample about half of people in the effective altruism movement in 2013 were no longer involved five years later. If you think your current self is correct to be altruistic and don't want to leave donations up to a likely less-generous future person then bunching donations over several years is harmful: the substantial possibility that you don't actually donate outweighs the tax savings. [3]

(Thanks to someone I talked to at the SSC Boston Meetup for asking me the question that got me thinking about this.)

[1] Specifically, unless you had other reasons to itemize your tax deductions, you would do better to take the $12k standard deduction.

[2] This ignores inflation and investment income, but they aren't large enough to change the picture much over a ~6 year window.

[3] Depending on your views on discount rates ("how much better is it to give this year than next?") it might also not be so good.