This is an excellent list of how to evaluate scientific evidence.  Almost all the issues are applicable to evaluating whether a manager or investment approach will demonstrate outperformance in the future.  To wit:

• Differences and change cause variation – that outperformance is subject to a very high level of sheer randomness.
• Bias is rife – particularly given the huge rewards to finding (or claiming to find) an edge.
• Bigger is usually better for sample size – a 3-, 5-, or even 10-year track record is essentially meaningless.
• Correlation does not imply causation – self-explanatory.
• Regression to the mean can mislead – just when a manager or effect looks great (or terrible) performance will become much more typical.
• Extrapolating beyond the data is risky – it may have worked before, but the future is out-of-sample.
• Beware the base-rate fallacy – given the rarity of true alpha, historical alpha is probably meaningless.
• Controls are important – and you generally don’t have a control in investing.
• Randomization avoids bias – and again you generally don’t have this in investing.
• Seek replication, not pseudo replication – rolling periods are not separate tests because of the overlapping data.
• Scientists are human – and product providers are even more prone to bias given the enormous incentives.
• Significance is significant – if you try 20 things, one, on average, (by sheer randomness) will appear to be significant at the 5% level.
• Separate no effect from non-significance – the sample sizes (track records) are probably too small to find the effect even when the manager or strategy in fact does have true alpha.
• Effect size matters – differentiate between results that are statistically significant and those that are economically meaningful (generally not a problem in investing, if it is significant  it will generally be meaningful)
• Study relevance limits generalizations – the market/economy/etc. isn’t the same now as it was in the historical data.
• Feelings influence risk perception – self-explanatory.
• Dependencies change the risks – we rarely have independent factors.
• Data can be dredged or cherry picked – and in our industry it will be not just dredged or cherry picked, but tortured to death.
• Extreme measurements may mislead – lots of things led to the performance, not just the factor you are focused on.

In my head I have had a hierarchy of items I disseminate:

1. Quick simple thoughts or forwarding of things to my coworkers via email.
2. If it is a little more interesting also share (via email) with my consulting clients. (These are other RIA firms that have me on retainer for my perspective.  If you are interested in this let me know.)
3. If it is an item worthy of even broader distribution, email it via my “financial professionals” list at the end of the quarter.
4. If I think it is a useful analysis of a topic, flesh it out into a blog post.
5. If I think it is a really useful analysis of a topic, do a whitepaper (pdf) on it and put it on our web site.

In other words, I have always felt that a white paper is more important and permanent than a blog post, which in turn is more weighty than an email.  I never felt most of my random musings and comments were blog-worthy, but rather reserved them for (ephemeral) emails.

But Michael Kitces has been bugging encouraging me for a long time to put the content of the quarterly Financial Professionals email on this blog.  So I have (finally).  That email goes out the 15th of the month (or the next business day, if the 15th isn’t one) after the end of each calendar quarter (so January, April, July, and October).  From now on I’m going to post that missive here too (on the first of the month following that date).

So, this being May 1st, here is the first one.  I have uploaded the ones back to the beginning of last year as well.  They were never intended to be permanently available on the web, but I hope you find them worthy!

Financial Professionals Winter 2017

Financial Professionals Spring 2017

Financial Professionals Summer 2017

Financial Professionals Fall 2017

Financial Professionals Winter 2018

Financial Professionals Spring 2018

John Bogle has cautioned, “Too many investors-individuals and institutions alike-are constantly making investment decisions based on the lessons of the recent, or even the extended, past.”

Here is a chart of the major asset classes sorted by the differences between the penultimate downturn and the last one:

Asset Class	8/2000 – 9/2002	10/2007 – 2/2009	Difference
DJ US REIT	21%	-66%	87%
Alerian MLP	59%	-28%	87%
R2000V	8%	-51%	59%
GSCI	6%	-49%	55%
R3000V	-18%	-54%	36%
R1000V	-19%	-54%	35%
MSCI EAFE Small	-25%	-57%	32%
Barcap Long	30%	4%	26%
R2000	-25%	-51%	25%
Barcap 5-10yr	30%	6%	24%
MSCI EM	-34%	-57%	23%
CSFB HY	-2%	-24%	22%
Barcap AGG	25%	7%	18%
Barcap 1-5yr	22%	8%	14%
MSCI EAFE	-42%	-55%	13%
ML Conv.	-27%	-39%	13%
3Yr TSY	23%	11%	11%
R3000	-40%	-50%	10%
DJ Broad Stock Mkt.	-40%	-50%	10%
R1000	-41%	-50%	9%
S&P 500	-41%	-50%	9%
3Mo TSY	8%	2%	5%
6Mo CD	11%	6%	4%
3Mo CD	9%	5%	4%
BLS CPI	5%	2%	3%
R2000G	-51%	-50%	0%
R3000G	-58%	-47%	-11%
R1000G	-58%	-46%	-12%

As you can see, owning REITS, commodities, small value stocks, and MLPs all were great in the dot com aftermath and fixed income did really well also.  While the S&P 500 was going down 41%, equal weighting of the four assets mentioned earlier would have been up 24%, and an equal weighting of the four fixed income indexes highlighted would have been up 27%.  Thus a 75/25 portfolio (pretty aggressive) of 50% S&P 500, 6.25% each in the 4 fixed income assets, and 6.25% in the 4 other assets mentioned would have been down 8.1% in the earlier downturn.  But in the more recent financial crisis it would have been down 35.7%.

Suppose a client is going to contribute $X per month.  A trade could be done every month, incurring a transaction fee, or every other month, or every third month, etc.  When is the optimal time?

It depends on the spread of expected return of the asset class to be invested in over the expected return on cash (that isn’t really right, because they actually ought to be risk adjusted returns, but I am going to ignore that for simplicity), the amount of cash, the expected next inflow and amount, and the transaction cost.

For example, assume a client is going to add $1,000 to their account each month, the cost of the trade is $10, the expected return on cash is zero and on the prospective investment 8%.

The first month there is $1,000 in cash and investing it will cost $10.  That is a 100 bps cost but the expected return (monthly) is just 64 bps [1.08^(1/12)-1] so that is a bad deal.  The next month there is $2,000 and investing it costs only 50 bps, but the return for a month is still 64 bps so now it should be invested.

Here is the formula:

((1+I)/(1+C))^(T/12)-(T/D)-1

Where:

• D is the dollar amount of the proposed trade
• T is the transaction cost of the trade in dollars
• C is the expected annual return on cash
• I is the expected annual return on the proposed investment
• T is the time in months until there will be additions to the account or a rebalancing, etc. (I am assuming these changes will be of significant size.  Obviously, if the upcoming deposit is $1 it wouldn’t be worth waiting for.)

If the solution to the formula below is positive, do the trade, if negative, don’t.

Stock splits are an interesting topic.  There appears to be a widespread belief that a stock is somehow worth more after it splits and a somewhat related myth that lower-priced stocks have less risk and more return.

The best way for an individual to think about stock ownership is as if they were buying the entire company.  If you were going to buy the entire thing what would it be worth?

Suppose ACME is worth $1,000 and there are 100 shares outstanding.  Each share is obviously worth $10.  If the stock is split 2/1 there are now 200 shares each worth $5 because the company is still worth $1,000.  The investor’s position is unchanged. If the stock split 10/1 instead of 2/1 there would be 1,000 shares each worth $1 and many investors might have the mistaken idea that it is a great investment now because it appears so cheap (“It used to be $10!”), but the company has exactly the same value, in total, as it did before.

Another (frequently used) analogy would be cutting a cake.  Cutting a cake into more slices does not mean you have more cake.

Now, there are three things about splits that are positive though:

1. If the stock has a very high price per share, then splitting it will increase demand because more people have the wherewithal to buy it.  This is rarely true (even Berkshire Hathaway has “B” shares that are affordable).

2. Management doesn’t split the stock if they think it might go down.  They will be pretty certain that the business is going to do well (this applies to dividend increases also) before they make that announcement.  Thus the market, appropriately, takes the announcement as a very good sign for the company’s future.

3. There may be people who mistakenly (as explained above) believe it is a better deal post-split because the price is lower and they may purchase the stock driving up the price slightly.  To the extent they do this, the more knowledgeable market participants will be selling at the “too high” price (or possibly even shorting the stock).  In an illiquid stock though, the ignorant masses (so to speak) could move the price.

Now despite those reasons that a split is positive, there still isn’t an investment opportunity because the price will move on the announcement not on the split (actually it will begin moving on the anticipation of an announcement).  Stock prices (actually all securities prices) incorporate known information which certainly includes things that have been announced.  So the announcement from management is a good signal, but it is the announcement, not the actual split that moves the price (to the extent it wasn’t already anticipated).

Let me explain that anticipated part as well.  Using our earlier example ACME is “worth” $1,000 with no split announcement.  If there is a split announcement, analysts, investors, etc. will raise their expectations for the future of the company and value it say at $1,100 (essentially non-public information – management’s strong belief in the future – has become public).  Suppose there is a 90% chance that an announcement will be made. (Obviously opinions will vary as to probability, current value, value after the announcement, etc., but assume these are the consensus estimates.)  In that case the stock will already be trading at $1,090 ahead of the announcement reflecting the 90% chance that it is worth $1,100 and the 10% chance it is worth $1,000.  (Obviously in real life there are more than two possible futures and each one of those futures has an attached probability, but the concept is identical.)  You can only profit as an investor by having a better estimate than the average of all market participants.  This is extraordinarily hard to do.