Your IPS: saving you from yourself (part 2 of 2)
You've determined your goals. You've got a good idea of your risk tolerance -- at least, you think you do, and you'll no doubt have a chance to test that in the next downturn. You've determined your desired asset allocation, maybe set aside some money for speculation (or not!), and you've started to figure out your implementation. You might need another pass at that last part, though, because next up is asset location, and the asset classes assigned to your accounts may affect what you use to implement them.
First: your accounts should be as simple as possible, and no simpler. Do you have three brokerage accounts, all of which are (now that you've done your goal analysis) supporting retirement? Perhaps it would make sense to combine them into one. Do you have one savings account that serves as your emergency savings and your vacation fund? Consider splitting that account into separate targeted accounts (unless you use accounting software like YNAB to track each dollar's "job").
As you age, it becomes more and more likely that you have multiple investment accounts which simply cannot be combined. A 401(k) (Roth and Traditional), IRA (Roth and Traditional), each of those for your spouse, maybe a brokerage account -- it can add up quickly. How do you determine which asset classes go in which accounts? You could just mirror your asset allocation across your accounts, making them "clones" of each other as much as possible -- and this would likely put you in the top tier of investors, in terms of total portfolio organization.
But you could do better. There are tax optimizations you could perform -- which asset classes would do best in your tax-advantaged accounts, and which are the most tax-efficient and thus can safely be put in your non-tax-advantaged accounts? And perhaps some of your accounts (e.g. your 401(k)'s) have access to superior mutual funds not available to your retail IRA...or perhaps their only options for certain asset classes are poor compared to what you could find elsewhere.
Also, bear in mind that any sales in your non-tax-advantaged accounts will have tax consequences! Now, they may be positive or negative, depending on how the market has done; regardless, don't ever sell from your brokerage account without checking the tax consequence first.
This is an analysis worthy of its own blog post. For now, if none of this analysis sounds appealing to you, don't fret. As I said before, mirroring your allocation across your accounts, while not 100% optimized, is often still a very good idea (barring the aforementioned potential tax consequences). You could start there and revisit later.
Speaking of revisiting: once you've created this allocation, you need to maintain it. Imagine you created a 60/40 stock/bond portfolio back in 1977 and never touched it. According to Morgan Stanley, that portfolio would have fluctuated between 45% and 74% equities between 1977 and 2003. You would have found your level of risk varying greatly -- without your permission!
Periodic rebalancing is the best-practice solution to this. There are two primary methods of rebalancing: time-based and error-based. In a time-based rebalance, you simply revisit your portfolio on a periodic basis -- perhaps annually, perhaps quarterly or even daily -- buying losers and selling winners to bring your portfolio back in line with your allocation. In an error-based rebalance, you check up on your portfolio on a similarly periodic basis, but only rebalance a given asset class if it is a certain percentage away from its target.
So what is the optimum method and frequency? The answer, of course, is that it depends. I'll just point out the following:
Transactions are rarely free -- in most cases, the more buys and sells you make, the more you're paying your broker.
In a study from 2010, Vanguard determined that altering the frequency of a time-based rebalance had very little effect on standard deviation and returns.
In a study from 2007, researcher Gobind Daryani found evidence that the optimum threshold for error-based rebalancing was 20%. (You can get access to the study through membership to the Journal of Financial Planning.)
Finally: your Investment Policy Statement, while solid, should not be set in stone. There's a mantra that we battle-scarred engineers have learned the hard way: "premature optimization is the root of all evil." As I've mentioned many times in this article, your IPS doesn't have to be (and can't be) perfect the first time, and the perfect is the enemy of the good.
However, I do strongly recommend you set up a periodic review of your IPS, perhaps annually. Be diligent about it -- put a recurring reminder in your calendar program of choice! Create a list of questions to ask yourself in that review, such as the following:
Am I following my IPS? (If not, fix that!)
Have my financial goals or situation changed?
Has the external environment -- tax code, investment research, fundamental federal policy, etc. -- changed? (Note: current market conditions don't count. We're looking at long-term changes, here.)
Is there an opportunity for me to optimize my IPS? This includes adding complexity, simplifying, or both, depending on the situation.
Has your tolerance for risk gone down? (Don't ask if your tolerance has gone up. In a good IPS, even if you think your appetite for risk has gone up, you only change your risk tolerance towards the more conservative side -- otherwise, it's all too easy for you to increase your risk when the market is up and decrease it when it is down, which is precisely what leads to the "behavior gap" quantified in DALBAR's annual Quantitative Analysis of Investor Behavior study.
Have any other changes occurred in my life that might warrant an update in my IPS?
Like a good financial plan, an IPS should always be a work-in-progress. Life doesn't stop changing; your plans shouldn't, either. And while a good plan won't anticipate everything, it will give you guideposts to keep yourself from going off the rails.