1. You never sell stocks when they are down (multi-year, starting at retirement).
2. You sell stocks when they are an inflation adjusted 120% above their value at retirement. The 20% allows for some help from momentum. See the link When do YOU rebalance
3. You never buy stocks. Although it worked OK in 2008-2009 (and I did some rebalancing from bonds to stocks) what if the market is down for an extended period (a la Japan)?
4. Prime Harvesting backtests well (has a low failure rate) in the US market.
5. Prime Harvesting backtests well out of sample (UK and Japan).
6. When stocks are flat (inflation adjusted) you are taking withdrawals from bonds. This increases your % stock allocation. However, over time the expected return for stocks has increased. So your % stocks has increased at the time that stocks expected return has also increased.
What has me concerned:
1. Can have sizable changes in asset allocation. It looks like that is the price you pay to get a lower failure rate/higher SWR.
I'll quote myself from the book threadPage 87 wrote:Although it’s tidier and more comforting to have a fixed stock-bond ratio, it comes at a cost. The data indicates,
allowing income-harvesting strategies to float the ratio produces a substantial income premium. If the stock returns
are up, the bond percentage will typically rise as stock gains are locked in; if the stock returns are down, the bond
percentage will typically reduce to insulate stocks. However, the average is still dependent on the initial rate. A higher
initial withdrawal rate results in a higher stock average; likewise a lower initial rate results in a lower stock average.
I must say that even though I still frequently read new personal finance books, and read many posts here, it is really rare that I come away with anything that changes the way I invest. The last time that happened was reading a book from Dr. Bernstein where he does not recommend using bond ETFs since buy/sell spreads can significantly increase when the market is volatile. That is exactly the time you may want to sell for expenses in retirement. That lead me to reduce, and eventually eliminate, the Vanguard ST TIPS fund via ETF (and invest instead via mutual fund).
I've read many posts here on alternative withdrawal strategies during retirement. I was thinking of a fixed asset allocation (50/50) where I would maintain this allocation in retirement by taking withdrawals from either stocks or bonds to rebalance. I also was thinking (not fully decided) that I would not sell bonds to invest in stocks to rebalance. This was difficult for me during 2008-9 (I rebalanced a couple of times then stopped by selling bonds buying equities), so I can only guess how difficult it would be in retirement. I'm always keeping Japan in mind. No one is guaranteeing a quick market rebound to a deep bear market.
Reading chapter 3 in Living off Your Money did open my eyes to alternate ideas. I like the plan suggested since it is relatively simple to implement, and has done very well when compare with other withdrawal methods not only in the US, but in Japan and other places. The revelation to me is instead of a fixed allocation it is suggested to let the asset allocation float, but never buy additional equities once in retirement (allowing for reinvesting Cap Gains and Dividends). For equities you only sell when they are up using a one sided rebalance percentage. And by "up" I don't mean up this year. Up means up compared to the inflation adjusted stock balance at the time of retirement. So up covers multiple years, which I think makes more sense. Also, this method reduces the Japan issue, IMO, by not continuously catching the falling knife as it drops.