Thank you all for the differing answers. I now understand bonds a tiny bit more, but I don't really understand how a bond fund stacks up against a long term bond. I like to keep things relatively simple so a bond fund is enticing but I dont know what I'm missing out on buy not buying individual bonds.
99.9% of people should just buy a bond fund like BND. The most important factors to shop around are:
1) What it's invested in. E.g. treasuries, corporate bonds, municipal bonds, junk bonds, etc. These affect the credit quality which starts to matter in recessions.
2) The "effective duration" which is defined
here.
3) The expense ratio, although this is a minor factor.
Back in November, I specifically targeted some individual bonds with characteristics which would make their values rise fastest in the event of a decrease in the market's yield. So I was looking for treasuries with the absolute highest possible duration and the lowest possible coupon payments. Some of the individual bonds I bought yielded 1.25% when originally issued for 30 years, and I bought them at a steep discount. Others yielded 0% and had over two decades to go until the owner received $1,000. They performed exactly as expected as long-term rates collapsed.
I also bought high-duration bond ETFs with my spare change and interest payments, because you can only buy individual bonds in increments of $1,000, and sometimes $5k or $10k. I used morningstar to identify the funds with the highest effective duration, because that was my thesis at the time. (It is not my recommendation today)
So if you don't have a specific gamble like this, just BND set and forget. In hindsight, I could have accomplished most of my objectives with ETFs and probably paid a lower bid-ask spread.
Picking individual bonds might also appeal if you are an analyst looking to get more yield for less risk.