Assume everything else is flat, but the Generic Bond and Small-Cap asset classes are both down 2.5% each. In order to bring your total allocation in these two asset classes back up to normal, you would take $2,500 and buy more of the Generic Bond's mutual fund, and also add $2,500 to the Small-Cap mutual fund.

You are now back to the original Asset Allocation Weights, and this is called rebalancing. One of the best advantages in the methodology is that it forces you to eliminate all of the emotion, market timing, gut feeling, intuition, tips, and all of the other things in the investment world that end up losing more money than it makes.

It forces you to sell some of the asset class that has went up recently, and at the same time, it forces you to buy more of the asset classes that have went down recently. The #1 saying in Wall Street is to "buy low and sell high" - and this is the best methodology ever invented to be sure you're always doing that, without having to fret over it. In other words, just do it, when you're supposed to, regardless of how you "feel" about it.

In reality, since the dollar amounts will be smaller for some people, it may be impractical to buy and sell mutual funds with these small amounts. But since models are never Real World, the software (the Linked Returns spreadsheets you get that calculate the actual returns) mechanically rebalances at the end of every quarter, even if only a few bucks moves around.

Again, models are rarely Real World, which is why they're called "Models." The point is that even though there are numerous practical limitations, it's still one of the best ways to manage money - if you're a consumer or a professional. The returns of all three models are shown on the main model pages, and sort of in the demo. You can see that they consistently outperform the market indices, which is the point.

Even though this may seem like a strange way to do things, it is very old-hat in the investment world, is taught in every real investment school, and is practiced by the majority of financial planners and professional investment managers (although not very well, and most financial consultant's asset allocator models are very mismanaged and therefore are very misleading, meaning you're not going to get the returns they showed you).

So if you're using the model allocations, you very much need to pay attention to rebalancing to get most of the benefits of asset allocation.

If you bought support, then you can ask questions at any time. The best questions are from people who ask if they should switch mutual funds and rebalance when they get their monthly mutual fund picks and investing model updates. You can send details of your situation and you'll get a Real World answer back.

Tip: Since thousands of investors are rebalancing at the same time, you may want to pick a different day to rebalance than the first day of the new quarter. You may make a few extra bucks by doing it a week before the new quarter. That way when people are selling things that went up, you'll be selling before everyone else, which means there will be less downward pressure on prices. And if you're buying asset classes that went down, doing it before everyone else will allow you to buy before everyone else, again beating the upward pressure. So the best time to do it is as soon as you get your updates in December, February, April, September.

Making Sense of All of the Different Return Numbers

There are several types of returns listed, so this section helps explain them. If you still don't understand, then please e-mail with an explanation of what you don' understand, and things will be edited until everyone gets it.

Around the 20th of every month, the Mutual Fund Picks, Models, and the Portfolio Statistics sheet of the Comprehensive Asset Allocation software are updated. Updates are sent to subscribers, and then new returns are posted on the site.

First, terms explained:

Hypothetical Returns: Returns shown on the spreadsheet "Models for Month Year.xls" (that you get when you buy them), the model portfolio's "demo," and those shown on the first row of the table on the main asset allocation page, are all "hypothetical." All this means is that the returns shown are just using the past returns of the current mutual fund picks. So accounting for past trades/holdings is not accounted for.

As you can see by comparing the first and second row of the table on the main asset allocation page, Hypo returns are usually much higher in shorter time frames (less than three years) than actual returns. This is because fresh picks usually have better recent performance.

Most advisors using their own models are usually showing these kinds of hypothetical returns, so judging performance using these numbers can be VERY misleading. The reason they do this is not because they want to show higher returns to prospects than they've really been getting, but because they're not using expensive portfolio management software that calculates returns accounting for past trades, and/or because they're too "lame" to do the work needed to calculate actual returns manually (like we do).

There's nothing wrong with using hypothetical returns. Because of the price tag of portfolio management software, and the dozens of hours of tedious work needed to calculate actual linked returns manually, the vast majority of investment managers use them. So BD/FINRA Compliance people allow this, as long as all of the disclaimers are there. It took several month's work and hundreds of dollars getting our models approved for use by a BD, FINRA, and the SEC in 1999. After that, they sailed through the next BD's compliance people without problems in 2001.

The graph at the very bottom of the Index Model (also shown on the demo) shows how the models have done when properly compared to their benchmarks. The yellow bars show the difference between returns of the models vs. the same model funded with indices.

Hypothetical Returns Before and After Fees: When you see "fees" in all of this, it means investment management fees that a professional advisor would charge their clients. This is not an extra charge we as the model providers charge you for doing it yourself. Individual investors using these models for investing their own money do not pay these fees, so you can ignore them all. The average fee is ~1%. Internal mutual fund fees (management and 12b-1) are already accounted for in the returns, so no adjusted is done for that (mutual fund returns are always after all internal mutual fund fees and expenses). So when you see fees here, these are not referring to internal mutual fund fees.

When we post after-fee returns, it's most always with a 1% advisor fee, and a 0.1% in average trading costs (ticket charges). So we most always subtract 1.1% in fees.

After-fee returns are posted (and there is a place to input them into the models) because it gives advisors a way to compare to other money management systems, and to show clients before and after fee returns. Advisors know what this is all about, and again, individual investors can just ignore it all.

So when hypo returns are after fees, this means that if the model got 10% and the fee was 1%, then the Last 12 Months (and/or Last Three Years) returns will be 9%. This 1% fee is prorated monthly (divided by 12) and subtracted from the Year-To-Date returns. So 0.5% will be subtracted from the model's returns if YTD if six months went by.

Actual Returns: Returns accounting for all past mutual funds held (which are not being used anymore and are not now in the models), are typically called "actual returns." They are also called "linked returns" because the return of the old fund is manually linked to the return of the new fund. In hypo returns, there is no linkage, as the new fund is just assumed to have been held all along (bad!).

This linkage makes the past returns as realistic as possible, and much more accurate than hypothetical returns. So these are the most important numbers.

Linking returns uses the "Time Weighted" methodology, and is the standard used in accounting for past trades in computing historical investment performance.

For example, if ABC Fund had decent returns for five years, and then started getting dismal returns, it would be replaced by XYZ Fund - the current hot fund. So just by replacing that one fund, overall hypo model returns can easily be a third higher. These higher returns are then shown to potential investors, even though ABC Fund was used up until the last month. So all one needs to do is replace one fund, and then use the historical returns of the new fund, and their track record has been substantially "faked" to make the manager look way better than they actually are. So the new investor would think they bought into a manager that was getting an average of 10% over over the last five years (because XYZ Fund's returns are being used in the hypo results) when really they didn't hold XYZ until last month, and so the results he got from ABC would have resulted in only a 7% average return. So as you can see, this is very misleading and can distort the truth substantially. Always ask to see actual returns in addition to hypo. Just ask, "Are these returns adjusted for past trades, or is what's shown just the past returns of the current holdings?" More than likely, they will be the latter, which is bad.

It takes a lot of care and work to maintain actual returns, so that's why it's rare to see them posted from an investment advisor. Actual returns will most always be lower than hypothetical, so the advisor has little incentive to calculate them accurately (also because their BD and FINRA will let them use hypothetical). The vast majority of posted investment returns are hypo because investors are usually clueless about this very important distinction, and thus don't demand it. So some good advice is to try to avoid managers using hypothetical models if possible, as the differences can be enormous (sometimes almost double). To simplify things, subtracting a quarter off of all hypo returns will get you into the reality ballpark (if a manager is showing 10% hypo, then it's probably only 7% actual, which is an enormous difference - the difference between steak and dog food during retirement!).

How we account for actual returns in our models:

The actual returns are from inception (1 January 1999) showing the returns as if you invested on the first trading day of 1999, then made no more deposits nor withdrawals, paid no taxes, reinvested all capital gains and dividends, rebalanced on the first trading day of every new quarter and when allocation weights changed, and switched all of the funds on the first trading day after the switch was announced. This never happens, and can't, so even actual returns are hypothetical. Even so, the grand distortion, lack of accounting for past holdings is removed; making things as good as they're going to get.

Our actual returns are generated using three spreadsheet that you get when you buy the models:
Fee-Based Linked Model Returns.xls
No-Load Linked Model Returns.xls
All Load Linked Model Returns.xls

These are all locked up but transparent, meaning you can see the formulas so you can see exactly what's going on, so you can do an audit yourself, but you won't be able to change anything (you can only change the amount of assumed fees in the Fee-Based Returns spreadsheet).

Miscellaneous

We get Morningstar data about mid-month, which means everything is two weeks behind already (e.g., April data isn't received until mid-May as it takes time for them to compile it all and send it out to customers). We typically do the work of screening funds and updating models the next day (and updates are sent out the next day).

So when a fund changes, it wouldn't be "right" to assume that the fund was held at the beginning of the month (because it wasn't selected until mid-month). So the previous fund's monthly return is used in the calculations (and is shown as gray in the linking spreadsheet). This delay is where the linking takes place. So if you see a fund pick in April that's new and gray, then the previous fund's return was used in the calculations here (not the brand new fund's return).

So in mid-May, you'll get the models with April data. The new fund picks are assumed to have happened at the beginning of May (end of April), which is already two weeks late. So yes, the data you're getting assumes you made the trades at the end of April, but you didn't get your update until mid-May, so you are correct in thinking that everything is delayed two to three weeks. There's noting that can be done about any of this because we can't get data sooner. Don't despair though! Based on observation, the chances of getting higher returns are about the same as getting lower returns even after this delay. In other words, people whine that their fund picks are not able to be implemented for two weeks after they are assumed to be accounted for in the models. The knee-jerk assumption is that money was lost here because you didn't get a chance to sell the old bad fund and buy the new good one for two weeks. In reality (we tracked this for a few years to see) the odds are just as good that you'll make money as losing it here, as share prices fluctuate all over the map at random during this two week period. So don't worry about the delay, chances are 50/50 that you'll make money and not lose any because of this delay. This is just the way it is, it can't be changed, and everyone else that does this is doing the same thing. So now that you know, please don't whine about it!

Actual Returns Before and After Fees: When investment manager fees are deducted from the model in actual mode, it's different than hypo mode. In hypo mode, they're just subtracted from the gross return (10% - 1% = 9%. The whole thing is a fudge anyway).

Fees are deducted in actual mode automatically on the Linked Returns spreadsheet (only on the Fee-Based Models. No fees are deduced from the Load nor No-Load models because this would be rare). You just go to the far left sheet and input them, and everything is automatically adjusted. Return comparisons shown on the table on the main asset allocation page are all adjusted for assumed 1.1% annual fees (which is average in reality).

Fees are accounted for in a way as close to reality as possible (billed in advance quarterly). At the beginning of every quarter, annual fees are divided by four, then applied to the gross account balance. So if the fee is 1%, and the account is worth $100,000, then $250 is the quarterly fee (1% divided by four). This $250 is deducted from the cash/money market account (row 35), before returns are calculated, so this debit is taken into account. If there is no cash, like in the Aggressive Model, then it's assumed to be taken from the safest bond fund.

There are no mutual funds with redemption fees (B shares) used in any of the model allocations.

As you can probably see, we're doing everything possible to have the posted returns be as accurate as possible - way way way more than anyone else.

If you maintain your own model allocations, you can also hire us to calculate your true historical returns by linking all of the past trades.

Here's Where the Returns Are:

You can see the hypo returns of everything on the far right sheet of the models (after you buy  them) on the sheet tab All Returns at a Glance. This is also pasted into the model's "demo."

The only places you'll see hypothetical returns are on the spreadsheet "Models for Month Year.xls" (that you get when you buy them), the model portfolio's "demo," and those shown on the first row of the table on the main asset allocation page. Everything else is actual (and they're labeled in the fine print).

All of the usual, and useless, portfolio statistics (Sharpe, Beta, Alpha, etc.) for the Moderate Fee-Based Model are on the third to far right sheet of the Comprehensive Asset Allocation software demo.

To see the latest actual returns, go to the bottom of the last month’s area in the latest year’s sheet tab in one the three Linked Model Returns.xls spreadsheets.

These three files don't do anything for your clients and are just FYI. This spreadsheet is protected, but the cell formulas are not hidden, so you can see and audit the formulas and returns. This is also where the history of all of the fund changes are. Input your hypothetical asset management fees into the far-left sheet of the Fee-Based spreadsheet.

Actual returns of the five Fee-Based Models, without fees, and market returns are on the table mid-age here: http://www.toolsformoney.com/model_portfolios.htm

Actual returns of the five Fee-Based Models, with fees are here: http://www.toolsformoney.com/model_portfolios_with_fees.htm

Actual and hypothetical returns of just the Moderate Fee-Based Model, compared to Index Models, the markets, the American Funds model, DFA,  SEI, and Warren Buffett's BRK-A are on the table mid-age here: http://www.toolsformoney.com/asset_allocation.htm

Hypothetical returns of each asset classes’ mutual fund pick, compared to its benchmark index are on the table mid-age here: http://www.toolsformoney.com/mutual_fund_recommendations.htm

Hypothetical returns (not accounting for past trades) of all of the models, with fees deducted, are on the model demo Word doc here: http://www.toolsformoney.com/model_portfolio_demo.doc

Actual returns (accounting for past trades) for the no-load models are here: http://www.toolsformoney.com/no_load_funds_investment_model.htm

Actual returns (accounting for past trades) for the all-load models are here: http://www.toolsformoney.com/load_funds_investment_model.htm

About Switching Mutual Funds when They're Changed on the Mutual Fund Picks Spreadsheet

First of all, one of the screening criteria is to weed out mutual funds that have back-end redemption fees, (B shares), so there won't be any back-end loads or fees to worry about when you sell mutual funds. The only thing to be wary of is capital gains taxes.

It's impossible to give exact advice here about when to switch a mutual fund, because there are too many factors to consider. The purpose of switching funds is to use a fund that will behave most like the asset class going forward.

The new mutual fund is not intended to "go up" the most. The new mutual fund is intended to act like the asset class the most, and at the same time, go up more than that asset class when that asset class is going up, and go down less than the asset class when that asset class is going down - over the next year or so. This isn't a Morningstar-type firm, where they get paid-off to tout a mutual fund (or mutual fund family) as being "the hottest thing since the invention of the wheel."

In general, it's usually best to switch the mutual fund in the Real World when a switch is made. It's also always best to not put any new money into the old mutual fund after the switch. However, there are numerous Real World obstacles when it comes to switching mutual funds.

So the first thing to do when you look at the spreadsheet Mutual Fund Picks.xls, is to look at the reason for the fund change. They are given in the comment sections of the cell, and display when you hover your mouse cursor over them.

If a fund was sold because it closed, then it's best to keep it. The reasons for funds closing usually are not bad. You sometimes won't be able to add brand new money to a closed fund, but most of the time, you can add money to a fund if it's already held before it closed. This is fine. Just be aware that the fund will be off the screening radar, so if it goes bad in the future, you won't be notified.

If the fund was sold because its style or asset class drifted or changed, then it's best to change the mutual fund.

If the mutual fund changed because of "performance," then you'd most always want to make the switch, as something bad happened to the mutual fund over the last month or so, and experience has proved that it won't magically fix itself, so the new fund most always does better going forward. However, the difference in performance may be so small that keeping the old fund is the most practical thing to do. The problem there is that the old fund is not monitored anymore going forward.

Mutual fund families (the overall mutual fund family name, like Oppenheimer) will most always pull shenanigans with their mutual funds if they think they can make more money. So they let the mutual fund managers change the stated objective of the mutual fund without changing the mutual fund's investing strategy. Fortunately, Principia's screening filter will catch most of these shenanigans. Mutual funds that use deceptive marketing tricks are most always screened out of the results.

The best reason to make a switch is when the trigger is performance related. But more Real World factors to consider are:

The tricks are to first pick one of the five investment models that match your investment risk tolerance (done via using the Investment Fact Finder), then sort your options by asset class, find the fund options that are not good, ignore them, pick only one good one per asset class, and then use that to fund the Investment Model's asset class.

It doesn't matter which model you choose to alter for your 401(k) (fee, load, no-load). Just remember that you only input returns and data into the Conservative Model (the other four models get the data from there).

The first step in doing this is to determine which risk tolerance category you best fit into. If you don't already have a financial tool for doing that, and don't want to guess, then the Financial Planning Software Fact Finders are here (you want the Investment Fact Finder). Once you know which of the five risk tolerance categories you're in, you can start using the Portfolio Model's directions.