Back when I was a young investor, like most self-confident, burgeoning people, I felt I could choose my own stocks.
Yes, yes, I know - you can’t expect old heads to rest on young shoulders, but nonetheless, that’s my excuse. Like most stock pickers, I had a general record of mediocrity punctuated by occasional flashes of success, or more often, highly entertaining failure.
My very first pick has haunted me for years. Being a beginner and a small-fry, I didn’t have a whole lot to diversify as I only had a small amount to invest. So I sank in my entire fortune of $1,000 into a UK stock that was trading on the New York exchange. Overall the stock had a solid record and great prospects. I watched it daily, awaiting imminent riches.
Instead, it tanked by about 30% and I lost $300 before reading a down forecast and cutting my losses by bailing out.
Later that year after selling my car, I came into some more money, about $10,000, and this too went into the same account. Now a little wiser, I put it into a diversified mutual fund. The market slowly crept up. In time, I had a 10% return - or about a $1,000 profit - so a net profit of around $700.
Strangely enough, my performance, as measured in percentage change since beginning this entire exercise, was still negative. It was initially -30%, followed by a 10% increase, which took me, when multiplying them sequentially, to a performance of -23%. Worse yet, it took several years more to make up that remaining 23% and get back to ‘even’. How could I have made money, but still have negative performance?
Welcome to the surreal world of performance measurement, where sometimes, if you’re not careful, up can be down.
In this case, I was using pure time-weighted performance measurement, which focuses exclusively on measuring the slope of the change of the assets regardless of their magnitude. Later profits, when the account was bigger, dwarfed the losses when it was smaller, but that initial plunge dominated my overall performance for years because I was not weighing it based on the size of the investment.
Performance measurement tools
Time-weighted performance is generally considered the gold standard of ‘pure’ performance measurement, except when its results turn out to be obvious nonsense. Just as a doctor uses different medicines for different conditions, different performance measurement situations can call for divergent procedural or analytical responses. For instance, here are 5 different scenarios and solutions to consider. Let’s start with two problems caused by common data discontinuities.
The ex-dividend price drop
Dividend lag time. This is a common scenario. Imagine an account with $100,000 market value invested in 1000 shares of stock XYZ, which is trading at a price of $100. The company has a hefty yield and pays a regular dividend. On March 20th, it declares that holders of record on March 21st will receive a dividend of $5 per share, payable on April 1st.
This means that starting on March 21st, anyone buying those shares won’t receive the dividend – that will go to the previous owner - so buyers obviously will drop their offer by $5 and only be willing to pay $95 going forward.
This drop in market price, equal to the amount of the dividend, means that the account as a whole declines from $100,000 to $95,000 in value. This 5% decline happens even for clients who didn’t sell their shares, but are simply waiting to collect (in this case) $5,000 in dividends.
In fact, the account has not truly lost value, some of the value of the shares has merely been converted into a pending cash payment. The check, so to speak, is in the mail - and on April 1st, when it arrives, the $5,000 payment in a $95,000 account amounts to a 5.3% jump in performance. April Fools.
Such phantom gyrations distort performance but add no real value to the performance measurement outcome.
Discuss with your tax and other advisors, the implications of accruing the dividend on the day it was earned and factor it into performance measurement, rather than waiting for the payment to post to the account and only counting it then.
The lagging split
A custodian is late in supplying the new share quantity after a split, but the price vendor accurately reflects the new share price on time. Imagine on Day 1 an account with 1000 shares priced at $100, for a total market value of $100,000.
On Day 2 the stock splits 2-for-1 and the price drops to $50, but the custodian quantity (perhaps waiting for DTCC) doesn’t post the 1000 new shares. So for Day 2 the custodian still sends us 1000 shares, but they are now priced at $50 instead of $100, for a new account value of $50,000, a -50% in the performance of the account.
Then on Day 3, the custodian posts the added shares, and we now have 2000 shares priced at $50, back up to $100,000 in total account value. The $50,000 additional value represents a 100% jump in the value of the position.
When needed and appropriate, compensate for the one day lag by correcting the market value and recalculating.
Data discontinuities can be managed with relative ease. More difficult are instances where the data is accurate and continuous, but just won’t readily lend itself to routine mechanistic approaches. Here are three more common scenarios:
Advisor management fee accounts
Advisor management fee accounts can have a zero balance on Day 1, $150,000 on Day 2 after fees have been assessed, and $-250 on Day 3 after advisors have been paid and an account holder who was accidentally over billed gets her refund.
With gains/losses of $0 on Day 1 and Day 2 the account will be performance flat, and on Day 3 trying to calculate performance of $0/-$250 will throw an error. Can anyone make a meaningful chain of performance returns out of data like that?
The best response may be to switch to using the Money Weighted Return method on a monthly basis or even to stop calculating performance for the account. It may be, properly speaking, out of bounds, so check with regulatory requirements and applicable industry standards.
Accounts trading on margin
Imagine a client who qualifies for margin privileges and uses them aggressively. She puts in $500,000 of cash, borrows $250,000 against it, and buys $750,000 of an index fund. Then the market goes up 5%, generating a $37,500 gain on the fund.
The account now has a net value of $537,500, reflecting a gain of 7.5%, not 5%. The whiplash effect runs the other way too: if the market had instead dropped, the account holder’s losses would have been magnified by the fact that her margin capability let her invest more in the declining vehicle than she could have invested without it.
This account holder’s performance can be seriously out of alignment with other clients putting their assets into the same asset. Some margin investors even want their margin balance designated as performance-neutral, blinding the performance measurement system to its effects, so that they can see the return of only their long positions. Not many performance systems do this easily.
Measure performance separately, perhaps in a designated composite and make sure you consult with advisers so performance is accompanied by required disclaimers about what is excluded and included.
The multiple management fee hit
One account within a household paying all management fees for the entire household is a common practice. Imagine a household with 3 accounts, each valued at $100,000 and each invested in the same index fund. The advisor fee is 1% per year or $750 per quarter for the entire household, but, because 2 of the 3 accounts are IRAs, the account-holder wants the fees levied against that third, taxable account. The index fund during the quarter goes up 5%.
In the IRAs that translates to a return for the quarter of 5%, but in the taxable account, the net of fees performance will be just 4.25% because the account is taking the hit for the fees of all 3 accounts.
Worse yet, if the fee account is radically smaller than the accounts for which it is shouldering the fees, it can endure severe negative returns on a net-of-fees basis, even while its meager assets are actually growing.
Grin and bear it, and perhaps make gross-of-fees returns available, if appropriate and disclosed as such, so they can also see the performance without the distortion. It’s what the account holder asked for. Just make sure the account holder is fully educated on the phenomenon.
These scenarios bridge the gap between science and the art of performance measurement. Numbers don’t lie - and when treated the same - should spawn predictable results, their proper interpretation requires experience and perceptiveness. Performance measurement calls for both the objectivity of reliable, repeatable formulas and the subjectivity of context and analysis that helps us create fair, ethical presentations that meet industry best practices and legal and regulatory requirements as a benchmark.
Investment performance standards
The Global Investment Performance Standards® (GIPS) can be an invaluable guide to ideal industry standards and best practices. They are a must-read for anyone who wants to know the ins and outs of fair, consistent, high quality performance measurement. AdvisorEngine®’s performance measurement system and data quality practices are designed with those high level standards in mind. It is not GIPS compliant though, that’s a process for you as a registered investment advisor to confirm.
In addition to generating daily exact time-weighted performance data in both net-of-fees and gross-of-fees forms, AdvisorEngine’s system accepts legacy data carried over from previous systems. Performance groups are user-configurable, permitting calculation of performance data for both households and individual accounts, performance data easily visible in graphic form on both the AdvisorEngine platform and in the client portal. More advanced features are on the way, including benchmark handling and Money Weighted Returns.
Accurate, trustworthy performance data helps cement long-term relationships between advisors and account holders by letting them know how their advisor has worked for them. It puts advisor performance into context and lays the groundwork for long-term mutual understanding. Experience it for yourself.
Investment performance is a heavily regulated topic to ensure investors do not receive and rely on inaccurate information. Before considering any opinions presented, please ensure you meet your applicable regulatory obligations for presenting performance to prospective and current clients before taking any action. This blog is sponsored by AdvisorEngine Inc. and CRM Software LLC. (“AdvisorEngine”) The information, data and opinions in this commentary are as of the publication date, unless otherwise noted, and subject to change. This material is provided for informational purposes only and should not be considered a recommendation to use AdvisorEngine or deemed to be a specific offer to sell or provide, or a specific invitation to apply for, any financial product, instrument or service that may be mentioned. Information does not constitute a recommendation of any investment strategy, is not intended as investment advice and does not take into account all the circumstances of each investor. Opinions and forecasts discussed are those of the author, do not necessarily reflect the views of AdvisorEngine and are subject to change without notice. AdvisorEngine makes no representations as to the accuracy, completeness and validity of any statements made and will not be liable for any errors, omissions or representations. As a technology company, AdvisorEngine provides access to award-winning tools and will be compensated for providing such access. AdvisorEngine does not provide broker-dealer, custodian, investment advice or related investment services. AdvisorEngine and Junxure are registered trademarks of AdvisorEngine Inc.