A problem is that declines in the stock market are correlated with mass unemployment, and there's a significant risk you'll lose more than $6,000 at the same time you become unemployed. This would leave you with an emergency fund of less than $20,000. If you were depending on that $20,000 to survive a period of unemployment, then you're going to have a problem.
This was actually taken into account in the article I linked to. Using their definition of a moderate-risk portfolio, looking at historical data it is unlikely that the account would drop below $20,000.
When deciding whether to give now or later, one factor to consider is the expected rate of return on money that could be invested to be donated later. However, for some people, having more money in investments may allow them to have a riskier overall asset allocation with a higher expected return, since they would be able to afford to lose more money if the investments decreased in value. This may increase the expected value of their later donations.
For these people, rather than looking at the expected rate of return of just the amount that could be donated now, it may be more appropriate to look at the difference in expected value of investments between a) donating now and having a lower risk tolerance and b) investing and having a higher risk tolerance.
By investing more, you may be able to do one or more of the following:
This does not apply to everyone. Some people's risk tolerances will be unaffected by adding more to their investments. Additionally, it might not be applicable if you expect to increase your spending, retire earlier, or give a larger inheritance to your family if you earn more from investments.
This is just one of many factors to consider when deciding whether to give now or later.
I am not a financial professional, so this isn't meant to be investment advice.
To illustrate how this might work in practice, the remainder of this post is a hypothetical, simplified example of the second bullet point above: how investing your emergency fund may increase your expected donations. All dollar figures are in today's dollars.
Say you have $20,000 in a savings account as an emergency fund. You want this to keep up with inflation over the long term, which you anticipate the savings account will do. If you use any of the money, you plan on funding the account back to $20,000 soon after.
Suppose you would be comfortable closing the savings account and instead investing 130% of your safety net, $26,000, in a moderate-risk portfolio. The excess amount would help ensure you had enough money available if the portfolio lost value. (This is the strategy recommended by Betterment.) You could accomplish this by investing an additional $6,000 that you were planning on donating.
If you expect an average of 3% yearly returns after inflation from the portfolio, the expected value at the end of, for example, 30 years would be $63,109, which would leave you $43,109 in excess of the safety net. If you think there is a 90% chance that you would donate this money 30 years from now (10% chance that you change your mind or your savings have been wiped out in a crisis), you would have an expected value of $38,798 in today's dollars to donate.
This represents a yearly compounding of $6,000 of about 6.4% after inflation (the percentage would be higher not adjusting for inflation). This is higher than the expected investment returns on the $6,000 because the asset allocation of $26,000 was changed as a result of saving $6,000.
At this point, you would have to decide whether it is better to donate $6,000 now or an expected value of $38,798 thirty years from now.
Interesting point. Basically your fixed emergency fund provides leverage.