|
| |
|
Six Steps For Successful Investing
|
| |
| |
| |
| |
|
|
|
Six Steps For Successful Investing
António Marques Mendes, Editor
|
|
|
 |
| |
|
| |
There are many platitudes about the art of investment. Some are truisms such as the saying that “investing is the art of buying low and selling high”. Others attempt to define investment with simple half-truths as when we declare that “investment is the art of choosing”. Even if we were to add more attributes to the definition by stating that “investment is the art of choice between competing assets” we would still miss the essence of the art of investment. Therefore, we prefer to create our own definition, designed to encapsulate all six steps for successful investing: 1 - Follow the markets; 2 - Choose a strategy; 3 - Be a careful stock and time picker; 4 - Build a well weighted and diversified portfolio; 5 - Manage the portfolio wisely; 6 - Regularly analyze its performance.
We define the art of financial investment as "a process of rotation between financial exposures which at the time of our choosing will allow us to attain the highest possible net liquidation value above that achievable without rotation and risk".
The breaking down of the investment process into its key stages forms the basis of our motto - “Every little bit counts and outstanding investors should strive to gain on each step of the decision process”. By utilizing this definition and following the six stages of investing, we are able to avoid relying on “monkey's luck”, and hoping for 10 consecutive heads while flipping a coin.
Choice is still the key component of our definition; however, the new definition encompasses various types of choice and not simply the choice between assets. For instance, by referring to exposures rather than assets we include both assets and liabilities. This takes into account the option to use leverage as a significant determinant of total return. Similarly, when choosing a specific liquidation date, it is important to consider the investment horizon pursued by different investors. Furthermore, when compared to the alternative strategy with no rotation (e. g. a strategy of buy-and-holding treasury bills) we are able to better incorporate the role of compounding in our assessment of investment returns.
We may say that our first choice is to 'choose whether we wish to choose[1]'. This is the question of whether we prefer a strategy of frequent choice over a strategy of choice minimization. Nevertheless, even if we choose the latter we must assess different risk appetites. The three main strategies for choice minimization – buy-and-hold, rule-based-investing and day-trading - are quite different when it comes to risk taking skills and aversion. For instance, a buy-and-hold investor wishing to bypass the need to choose which stocks to buy has the option to purchase a global or a single market exchange traded fund (ETF). However, the investor must still decide when to open and close exposure to the ETF as market timing may have an impact on returns. A day-trader can minimize choice by trading only one or two stocks. However, the day-trader is still left with a need to rotate between short and long exposures throughout the day. Clearly, none of these choice minimization strategies avoids the necessity to monitor the markets. We should therefore consider the choice between different market watching strategies as the first step on any investor’s checklist.
| |
Step 1 - Follow the markets
|
|
|
It is logical to choose an investment strategy before selecting the various ways to follow the market. However, from a practical point of view, one needs to review market information before being able to select the investment strategy appropriate for a particular investor's profile. Thus, we shall first consider the market monitoring process.
The frequency of market monitoring is a function of the timing strategy pursued and the scope desired. In order to follow the markets we must set up our investment barometer to measure the investment climate on a regular basis. To build an investment barometer investors need to do four things: a) choose a yardstick; b) construct the corresponding indexes; c) select a standard to convert all indexes into the same metric; and d) develop a barometer display so they can easily interpret market developments.
The selection of a yardstick involves the choice between monitoring the entire population of securities traded in a given market or, alternatively, a sample of securities. From a cost-saving perspective, the sampling approach is better, provided that we use tests of statistical significance or a consistent rule-based method of sampling. Since statistical sampling often requires very large samples[2], many investors prefer to rely on the latter option. Frequently very small portfolios are used as yardsticks, as in the case of the oldest and most renowned – the Dow Jones Industrial Average - which includes only 30 large capitalization stocks.
Historically, the selection of such portfolios has been based on the size of the constituent companies and more recently, on their valuation. However, we can widen the criteria we use to create rule-based portfolios. In our Newsletter and Investment Management Portal, we intend to make two new rule-based portfolio sets available – an Auto-PilotTM set of portfolios based on the best performing screening strategies for short term investors and another set of portfolios based on our StockMarksTM quality ratings, more suitable for long term investors.
When choosing from the increasing numbers of rule-based portfolio yardsticks investors have to balance three considerations – the popularity of such portfolios among the investment community, their statistical significance in replicating the overall market[3] and their ability to reflect the degree of volatility acceptable to the investor. Such choice is unique to each individual investor, but before they can make this choice they need to select or design a way to measure the yardstick’s evolution through time. This is usually done by referring to an appropriate portfolio index.
After selecting the initial sub-sample of stocks to use as the yardstick, the construction of an index only requires two key decisions – how to weight each stock and the definition of the rules for inclusion/exclusion of its constituents. Weighting may be based on one or more variables such as price, capitalization, net income, ratings and so on. Capitalization is the most widely used method.
Provided that all indexes have been constructed using the same method and are constituted by stocks traded in the same currency, they can be easily re-based to whatever date we wish. They can also be charted to give investors a tool to quickly compare various benchmarks. Indexes are calculated by the major index providers such as Dow Jones, Standard & Poor’s and Morgan Stanley and can be shown expressed in two or more currencies. However, converting an index calculated in a local currency into another major currency does not meet the requirements of a truly universal standard.
To develop such a standard we would need to take into account the purchasing power of each basket of stocks. This must be defined in relation to the individual objectives of each investor. An alternative to this individual approach is to make the assumption that investors will invest in the same markets forever and therefore measure the purchasing power of each basket in terms of all of the stocks the market. We used this approach in creating our own unit of account (virtual money) – the FEITM – which has an exchange rate in relation to all of the main currencies. This allows us to be as close as we can to an universal standard for measuring the performance of the various yardsticks used by investors.
Once we have converted all of our indexes into a common standard, we are ready to design the investment barometer. The investment barometer is a user-friendly way of displaying tables and charts depicting the evolution of our yardsticks. Often, the barometer may add some analytical indicators to its measurements, allowing us to assess the current market phase in the same way that barometers can show the seasons of the year. However, for the time being we will leave the discussion of the investment climate until the second step of the investment process – the choice of investment strategies.
| |
Step 2 - Choose a strategy
|
|
|
There is no single investment strategy befitting all investors. To define a strategy investors must begin by deciding which course of action will allow them to better tackle the four basic choices of investing – quantity of stocks, which stocks, their weightings and when to rotate positions. The decision process for these choices involves four main stages: a) identify the investor's profile; b) define the investment scope in terms of asset selection; c) set limits to risk exposure; and d) assess the investment climate prevailing in the market.
The first question an investor must raise is: what kind of investor am I? To answer this one needs to consider three key determinants – financial wealth, personality and objectives.
When investors evaluates their financial wealth, consideration must be given to both their current total wealth (excluding their residence) and the current allocation of their wealth between financial and non-financial capital, including how it is distributed in terms of asset classes. Investors should also forecast their future income and desired level of savings, as it is often said that to be a good investor, one needs to be a good saver.
There are three main determinants to an investor's personal profile. These include psychology, financial knowledge and time availability. To assess one's psychological personality one must to consider how he/she engages in maximizing and satisfying behavior. Furthermore, consideration should be given to one's ability to cope with stress, disappointment, greed and fear, as these factors influence competence in risk management and trading. Financial knowledge is also important in order to ensure that the investor fully understands the securities in which they are investing and the associated risks. Similarly, in terms of time management, investors should only utilize strategies requiring frequent trading when their profile suggests that they have the necessary willingness and time availability to sustain such a strategy.
When it comes to setting investment objectives it is important that investors clearly set their priorities. There is nothing wrong with people pursuing multiple objectives, but when dealing with the selection of investment strategies investors need to organize their priorities in terms of an overriding objective. Their objectives may be to maintain a regular income, to grow fast in terms of capital appreciation or to preserve the value of accumulated capital.
One does not need to be a specialist in combinatorics to realize that the multiple profile decisions described above can be used to define thousands of different investor types. Because of this, most investors may need to engage the help of professionals or profiling algorithms to identify their own type and more importantly to understand how profile matching might condition subsequent choices in selecting investment strategies.
The second stage in the characterization of an investment strategy is where an investor must define the scope of his/her investments. Investors must first consider the extent to which they want to manage their own investments. They can allocate between self-directed management (DIY investing), advisor-assisted investing or outsourcing through collective investment schemes (Funds) and wealth managers. Only after ascertaining how much capital they wish to invest personally should investors continue to select the appropriate decision making platform. For instance an investor with little capital or who regularly invests small amounts of money will need a platform such as our forthcoming platform DripInvestingTM. Other investors, wishing to outsource significant amounts of investment management through collective investment schemes need to select a product like our future STPFundInvestTM platform for fund selection.
The next step in the second stage of strategy definition is determining an investor's scope in terms of geographical macro capital allocation, asset class allocation and instrument selection. This process involves both allocation into investment accounts and markets. This is not only for for prudential reasons but also because unified managed accounts may not give access to all the desired securities.
Of course, the breadth of investment scope cannot be disassociated from the amount of risk individual investors can tolerate. This leads us to the third step in the strategy selection process – risk taking.
Risk tolerance assessment requires that we take into consideration both our risk profile and some of the major categories of risk. These include the market or price volatility risk, currency risk and leverage risk. Despite being difficult to manage, choices in terms of currency and leverage risks are quite easy to define. However, choices relating to an investor's risk profile and the market volatility risk can be harder to characterize.
Risk profiling requires that investors define their levels of risk aversion, risk awareness and risk tolerance in terms of extreme losses. Investor must also define a maximum loss limit. This limit can easily be set but will impact the investment scope due to the restrictions that it will place on the price volatility of securities for inclusion in the individual investor's portfolio.
The assessment of volatility levels requires that investors have clearly defined investment horizons either in terms of positioning in the market cycle (bull and bear periods) or in relation to the duration and rotation of their intended exposures. This leads us to the final step in the strategy selection process – assessing the investment climate.
Investors need to consider both the market environment they may face and how they will cope with, or take advantage of changing market circumstances. This can be done using both an investment barometer to assess the market outlook and an appropriate timing algorithm to assess the opportunities and risks of term timing.
Taking a market view requires that investors consider both the price volatility and the current momentum of the market, before formulating their thoughts in terms of market direction. It should also be remembered that taking a market view is not an attempt to forecast the future price of specific securities (this may be impossible) but only an assessment of the most likely direction and amplitude of changes in the evolution of the selected yardstick.
Term timing is a complex issue, and therefore not suitable for inexperienced investors. It requires the choice of relevant rotation terms and historical backgrounds before one can attempt to identify reversal signals for different long and short zigzags in the evolution of markets.
Finally, an important concern of investors throughout the entire strategy selection process is the need to observe the principles of portfolio diversification. These principles apply not only to risk management but also to strategy selection in terms of scope and timing.
| |
Step 3 - Stock and time picking
|
|
|
A very controversial issue among investors is whether picking stocks or timing terms will allow them to achieve higher returns than simply buying a fraction of the entire market [4] We have two opposing camps in the debate. One side advocates stock picking as the single most important determinant of returns and the other preaches that such activity is completely futile.
Among the latter we find the believers in an extreme version of the 'market-efficiency hypothesis'. This states that “prices on traded assets already reflect all known information and therefore are unbiased so that it is not possible to consistently outperform the market by using any information that the market already knows, except through luck”. The market-efficiency hypothesis translates into a price theory stating that stock prices and related financial series follow a random walk. Therefore, the determination of future stock prices is like a lottery process and whether investors choose their stocks by throwing darts or decide to select today’s best or worst companies is irrelevant. They are unlikely to generate consistently excess returns because the future prices of stocks will be randomly determined. The fact that brilliant minds[5] have considered this burlesque theory as more than a simple hypothesis derived for academic discussion is one of the most puzzling facts about how little finance theory has progressed in the last 50 years.
On the other hand, among the stock picking extremists we find the naive believers in the possibility of forecasting the equilibrium between supply and demand with great accuracy. This requires that we have enough resources to gather the information and computing time necessary to estimate well specified supply and demand equations for future prices. Stock picking extremists minimize the practical difficulties of the identification problem in distinguishing between supply and demand equations and the many difficulties associated with dynamic general equilibrium modeling.
Obviously, as is often the case, the truth lies somewhere in the middle. Seasoned investors dismiss both extremes and, while acknowledging that consistently beating the market is no easy feat, they also believe it is not only the domain of a few celebrated investors like Warren Buffett or George Soros. We believe that many investors can beat the market provided that they use appropriate decision tools and follow a disciplined and thorough path split into four stages: a) mapping; b) tracking; c) screening; and d) due diligence.
Mapping is the process of laying down the organization and search procedures required to implement an investment strategy. We would not expect prospecting companies to begin digging holes randomly in search of natural resources nor fisherman to throw their nets randomly in the oceans; similarly investors should also map their search field. In summary picking is a narrowing down process, which can be done in a top down[6] or a bottom up fashion as long as it is structured to translate the strategy adopted. In practice this means investors must select the right segment of the market and set up procedures tailored to take advantage of the best available historical database for each market segment and investment account.
There are two main tasks involved in mapping. These are data cleaning and setting parameters for the subsequent analysis. Data cleaning is the process of checking data availability, validation and date matching in order to solve the problems caused by outliers and information noise. Setting parameters involves the task of defining sub-samples of companies (or shoals as we will call them) and sub-periods for analysis. Shoal definitions may be based on a single dimension (e.g. sectors) or be multi-dimensional (e.g. industry, size, market correlation, style, etc.). As the number of dimensions and respective categories increases, the number of possible shoals grows exponentially and the investor ends up with such large numbers that can only be processed by powerful algorithms. This is why most investors limit themselves to only one dimension and a few categories (e.g. size is often split into three categories: large cap, mid cap and small cap). Even with powerful algorithms, we should be parsimonious and stick to a few dimensions (we set our limit at four) by privileging dimensions that may be popular among investors. If too many dimensions are analyzed, shoal tracking – our next step - may be very difficult.
Tracking shoals of stocks through several periods is indispensable when attempting to identify those with a higher concentration of winning stocks (or losers, for shorting) and a stable trend in terms of entries and exits from the shoal. The aim is to end up with a shortlist of shoals whose composition facilitates the process of successfully screening them. After being shortlisted by various methods based on their size and properties, the shoals should be also ranked in terms of the likelihood that they will have a reasonably high ratio of survivorship. Shoals with high survivorship ratios help us to identify potential winners by giving us larger samples of stocks.
Tracking should not be confused with back-testing, despite the fact that we can also back-test the selection of shoals. Tracking is a forward-looking technique while back-testing, as its name implies, is a backward-looking technique often used for accuracy testing. Tracking can not be disassociated from some form of term-timing and should be applied to shoals and not to individual fish as group movements are more predictable than individual movements. Having selected the shoal(s) on which we wish to focus our forecasting skills we can then move to our next step – stock screening
Filtering rules, derived from various sources and using different pass/fail tests, are the basis of any stock selector screen. There are a plethora of screens on the web but most of them are little more than an algorithm to query a database. A true screen must translate a proven strategy with a clear philosophy of investment and complementary rules to prevent side-effects and enhance results. The process of strategy design is probably the most creative of the processes in which an investor should engage. But he must do so in a consistent and thoughtful way and not by listening to rumors, tips and other forms of ad hoc advice; otherwise investors will move randomly in all directions like ships lost in the fog.
To design stock-picking strategies investors may draw inspiration from reading recommendations by reputable investors and gurus or by analyzing published predefined screens. In this regard, the consultation of screening typologies may be very useful, as they allow investors to quickly identify the kind of stock-picking strategies consistent their chosen investment strategy. Screens may rely on fundamental, technical or event data depending on the kind of database used, but it is crucial that they are designed to meet the investor’s need in terms of investment horizon (short or long term) and their desired exposure (long or short positions).
Having designed or selected the filters for screening the chosen shoal(s) investors should then select a number of stocks slightly above that which they need[7] to constitute their portfolios. The reason for this is that investors will have to to reduce their shortlist further by conducting a detailed due diligence with the aim of spotting and removing any rotten apples in their basket before continuing to prune it further through a careful diversification review.
Cleaning the list of selected stocks through a careful due diligence process is the final step of the stock-picking process. Due diligence can be carried out through detailed studies, through a scoring system, or through a combination of both approaches. Whatever the methodology utilized, due diligence should give the investor a clear picture of the each stock’s strengths and weaknesses.
To measure its relative strengths and weaknesses one must use an appropriate set of indicators to compare each stock in the list of selected companies with both a list of its peer companies and the rest of the list. Indicators to evaluate strength should include measures of historical returns, profitability, financial strength and market momentum. Likewise, the evaluation of weaknesses should be based on indicators of quality, valuation, price manipulation and trading and accounting scams. To be effective, due diligence needs to be conducted in a timely, regular and cost-effective manner. This requires the use of vast resources or alternatively, access to adequate due diligence tools such as our forthcoming OnlineDDTM.
Due diligence completes the stock-picking process and investors can then move to the portfolio building step of the investment process.
| |
Step 4 - Build a well weighted portfolio
|
|
|
Portfolio building is often done in one of two distinct ways – in a single major exercise or in continuous small steps. The first can be done either from scratch or by a substantial restructuring of existing portfolios, while the second is usually done through the drip investment of savings or at regular intervals by rebalancing the composition of an existing portfolio (for instance using our StockMarksTM quality rating). Whatever the case, investors should always build their portfolios on the basis of a previously created stock watch list.
Watch lists can be created through a structured process of stock picking as described above or by adding up stocks in a more or less ad hoc manner. We can create our list from several runs of our stock picking algorithms or by combining different strategies. No matter which route is chosen each stock should have be evaluated against their peers on six majors counts – payment and volatility risk, overall quality and due diligence scoring, as well as on its current momentum and outperformance likelihood.
The purpose of using watch lists is not simply to have a list of eligible securities, but to make sure that such securities have been chosen in a consistent and timely manner. For this they must be subjected to two types of tests – inertia and diversification.
Inertia testing involves the analysis of the frequency with which securities appear in recent screening runs (to avoid stocks that are stuck in a motionless state) or in monitoring securities during a probationary period to see if they meet our expectations.
Diversification tests can be based on similarity analysis (e. g. excessive concentration in the same sector, size, beta, etc.) or on the statistical analysis of the correlation between the stocks that constitute the portfolio. Given the statistical property of variance referred to below, the analysis of correlation should be the favored approach. Nevertheless, one must be always aware of its limitations because of possible spurious correlations and the inherent instability of stock correlations. In most cases we deal with only one or two dozen stocks and short periods of time. Often in such circumstances the substitution of a few stocks is enough to change the sign of correlation, let alone its value. After diversification and inertia testing we should end up with a list of enough stocks to invest our capital. Now we must move to the second stage in portfolio building – the capital allocation to each investment position included in our watch list.
Investors may choose between several weighting methods. The simplest option is to equally weight all stocks. A more complex method would be to use differentiated weighting, using criteria such as market capitalization, quality ratings, due diligence scoring, price volatility and some other forms of user-defined preference rankings. Such weights may be obtained by simple scaling methods or by optimization models.
Among the latter the most widely recommended is the mean-variance optimization model which relies on the fact that a measure of dispersion more often taken as the measure of risk - the variance – has the statistical property that the variance of a sum (e.g. the sum of monthly returns on two stocks) is smaller than the sum of each variance taken separately. Moreover, the variance will be smaller the less correlated the stocks are. This is the basis of the so-called modern portfolio theory, presented in 1952 by Harry Markowitz. By deriving the curve of efficient allocation depicting the optimal trade-off between return and risk, investors are able to choose an optimal allocation for their preferences.
Although mean-variance optimization is a technique recommended for capital allocation, it is also used occasionally as a stock picking tool by dropping those stocks with a lower or negative weight. Indeed the technique can be used with more variables than simply the monthly stock returns; we can also include our own quality ratings. This does not guarantee greater efficiency in stock picking but may help with the study of diversification for the construction of watch lists.
The major drawback of this optimization approach is that it relies on the availability of three types of forecasts which are unreliable and difficult to obtain – future returns, their standard deviations and correlations. Because of this, most professional investors tend to either disregard this method or alternatively, use historical values which are barely reliable for long historical periods and are therefore more useful for asset allocation than portfolio allocation.
Unfortunately, empirical evidence suggests that the various weighting methods referred to above tend to perform disparately under different market cycles but it is not possible to identify clear patterns. The best we can do is to suggest that investors compare their portfolio allocations under different weighting approaches before making a selection. Having finished the weighting of your watch list it is now time to plan how to trade in order to open the desired positions.
Trading can be done by investors or outsourced to a professional trader. Orders can be given one by one at different times or placed as a basket for execution during a certain time period. There are many types of orders ranging from simple limit and market orders to very complex stop and conditional orders. Inexperienced investors should stick to limit orders and avoid trading close to the daily opening or closing of the market.
However, for experienced investors trading planning and timing can be an important source of gain provided that they have the necessary time, skills and algorithms for technical analysis. As part of the process to secure the best possible execution of your trades there is also a need to keep good records and to check the settlement of your transactions. After your orders have been executed your portfolio has been created and you must now check if its allocation needs minor rebalancing due to significant differences between executed trades and your initially intended allocation. You must also prepare to follow your portfolio on a regular basis. This is usually done not by simply reading the monthly or quarterly broker’s statements but by using online monitoring tools.
Monitoring is usually carried out with the help of a portfolio monitor provided by brokers or by websites. Investors can also build their own portfolio monitor in a simple excel spreadsheet. A good monitor should include the possibility of following the performance of each trade individually while allowing the aggregation of trades by position, instrument, currency, account and so on. Facilities to import data directly from your broker and to chart the recent evolution of prices are also advantageous features, but the essence of a good monitor is how well it displays the relevant data for instant portfolio monitoring. This will depend on your personal taste but should always reflect the nature of your exposures. For instance, if you invest in leveraged instruments it should give you an immediate view of your margin requirements.
How often you need to check your portfolio depends on your investment strategy and its management requirements. These will be addressed at the start of the next step for successful investment – portfolio management.
| |
Step 5 - Manage the portfolio wisely
|
|
|
Portfolio management is a key determinant of investment success. No matter how careful we are at stock picking and portfolio building the market will always surprise us, either positively or negatively. Your performance is dependent on how well you manage such outcomes. However, before moving into how to decide on position management and closing, investors should begin by considering two preliminary stages – positioning analysis and the regular reviewing of their portfolios.
Positioning analysis was partially considered in the portfolio building step, but now it needs to be geared towards the management of when to realize capital gains and losses. This involves setting up two key tools – a risk management framework and an alert system.
Depending on the exposures taken, risk management may involve deciding on the use of currency overlays to protect against currency risk, defining the levels of leverage used and establishing a set of protective orders. All these tools imply that the investor has a clearly defined set of triggers based on unrealized gains and losses or on price targets.
Such triggers must be integrated with other signals based on news and events and put into an alert system capable of delivering timely information to the investors. Nowadays, alerts can reach the investor almost in real time no matter where he is, usually via email or SMS.
Reviewing is the assessment of results and trends that may affect the investor’s performance. Provided that investors have a good alert system, they do not need to spend their life glued to portfolio monitors, and only need to review their portfolios at regular intervals. These reviews should focus on the divergences between the initial expectation and the actual outcome, with a view to either rebalancing the current strategy or changing course by switching to another strategy. It is important that these two options be assessed with a cool head and as far removed as possible from recent events so that investors do not change course too frequently and erratically.
Apart from using a consistent method to examine risk adjusted returns during the evaluation stage, this review process should aim to classify stocks in three broad categories – those within expectations, those with extreme losses and those showing great unrealized gains. Of course such groupings must be decided on the basis of the historical volatility of the market and of the stocks in our portfolio. What might look like an extreme loss during bull markets may be considered a normal loss in bear periods. The ratios between these groupings provide a good basis on which to evaluate whether to rebalance or not; but fundamentally they should trigger the adoption of profit taking or loss recovery strategies. Reviewing should therefore be the foundation on which the managing stage rests. However, it remains dependent on timing considerations.
Managment decisions always relate to choices of whether to close outstanding positions and on what to do with the generated cash, but they differ in terms of each investor’s strategy. For instance, one expects that a long term value investor like Warren Buffett, who advocates buying a stock for life, will only close his positions after becoming convinced that he made a mistake in the first place or that the market has gone wild on the premium it is paying for the stock. In this case, one should engage in profit taking by temporarily reducing exposure to such stocks. On the other hand, investors with frequent trading strategies are likely to close any position regardless of its valuation as soon as their self-imposed holding periods expire. In relation to deciding what to do with excess cash positions, investors usually follow three main paths: hold the cash until the next review period, search for new opportunities outside their current watch list, or reinforce their existing holdings by favoring stocks in one of the three performance groups mentioned above. Again, some investors may prefer to reinforce their bet on the current winning stocks or on lagging stocks depending on their investment strategy.
Whatever the investor’s strategy, investment management decisions are always precipitated by two different types of events – extreme price changes or market and corporate events that force investors to reassess their valuation of the company. In the first case, having been surprised by extreme price changes, investors have to deal with the anxiety caused by fear or greed in order to take one of three courses of action – to hold to the current position, to increase the current position (e. g. double or triple betting), or to reverse course by either closing or reversing the current exposure. In relation to non-price events, investors must assess their likely disturbance to prices and most importantly the likelihood of such impact being transitory or long lasting. It is also important to remember that sometimes investors do not have the luxury of waiting before making up their mind like when a majority of shareholders votes in favor of takeover offers or share buy backs.
Nowadays for many of these decisions investors may make use of valuation and management tools like recovery and profit taking algorithms, but in the latter it is their judgment that must prevail and this must be continually trained in order to ensure that the number of correct decisions is greater than that of the inevitable mistakes. Once decisions have been made to close a given position and how the proceeds will be used, the next step is to translate the decisions into trading orders.
Trade closing skills are not fundamentally different from those needed to open new positions, except that now the objective is to get the highest price possible, not the lowest. Therefore, the recommendations made in the portfolio building step of the investment process about trading apply equally here. Nevertheless, a few aspects are more critical in the closing stage, namely the use of stop loss orders, block trading and getting out of illiquid stocks.
For better results with the latter it is necessary that you work with a brokerage firm that has good routing algorithms and access to first-rate pools of liquidity. As for the use of stop and other conditional orders it is fundamental that investors become familiar with their rules and that the brokers are honest and transparent in disclosing execution details. Otherwise, investors may end up closing their positions in less favorable terms than envisaged at the time of the decision.
While the regular reviewing of portfolios should be frequent (e. g. monthly or quarterly), investors should also define a less frequent (e. g. yearly) evaluation of their investment performance. This is the final step in the investment process and it should be designed as a self-learning tool.
| |
Step 6 - Regularly analyze the portfolio performance
|
|
|
As in the previous steps we shall distinguish four main stages in the performance evaluation procedure: a) the preservation of gains; b) the measurement of returns; c) the benchmarking of the results achieved; and d) the breakup of performance for return attribution and learning from experience.
The preservation stage in the evaluation process is primarily devoted to ensuring that the after-tax results are as high as possible, but it must also address the choice of adequate risk management limits and hedging. Tax planning should be taken into account at three different stages in the investment process – when deciding the capital allocation into asset classes and markets (or accounts), while choosing between strategies and while closing positions near the end of the fiscal year. Because governments keep changing the tax treatment of capital gains and income and often offer tax exemptions for selected investment vehicles or securities it is difficult for investors to keep track of such changes. Under these circumstances it is advisable that investors seek tax advice.
Another important method of gain preservation is to monitor transaction costs to make sure that investors do not fall victim to churning and use the most competitive intermediaries.
Finally, investors who regularly change the constituents of their portfolios, may preserve more of their gains if they decide to hedge all or part of the unrealized gains near to the closing of their positions. This is particularly true during periods of high volatility, but investors should also bear in mind that hedging is not a free meal and therefore assess its costs. After taking all the necessary precautions to ensure that investors keep most of their gains they must also measure such gains (or losses) on a regular basis.
The assessment of results is not as simple as calculating the difference between the current liquidation value of all exposures and their value before the portfolio was built. The main reason being that we need to report such results to a given holding period and to compare it with the no-rotation no-risk alternative. To do this we need to use standard periods of time (e.g. monthly or quarterly returns). Therefore, when we assess a period made up of several sub-periods of time we must consider some type of average return for the whole period and to take into account its standard deviation or other measure of mean-centered dispersion. In practice we find that although most investors choose to use a simple arithmetic average, more sophisticated investors may prefer to use a geometric mean. These measures can be substantially different.
Even when investors assess only single period returns, they still need to consider if such return was achieved with a higher or lower level of risk. Thus, they need to use the standard deviation as well as some measure of risk-adjusted returns. Among the most widely used are the so-called return-to-volatility ratios, namely the Sharpe and Treynor ratios. Since these are more often used to assess multi-period returns it is important that investors distinguish between portfolio volatility over time and the volatility of the constituent stocks within the portfolio during a single period. Also, because investors are not solely interested in comparisons with risk-free alternatives they must consider the benchmarking of their portfolios.
Benchmarking is the key to performance evaluation and it must begin by choosing the right kind of benchmarks. In fact investors should always use at least two benchmarks – a broader one based on index replicas of the entire market and a like-with-like benchmark based on indexes or funds within the same strategy category. The first are the most widely used, albeit using an index that is not truly representative [8] of the entire market.
The most widely used benchmarking statistics are based on a linear regression, between, for example, a monthly portfolio return and the corresponding benchmark index. They are usually presented as the R-squared (the correlation coefficient), the alpha (the intercept) and the beta (the slope). These values suffer from the usual caveats associated with linear regression, and investors should be particularly careful to avoid being misled when these parameters have very low statistical significance or were estimated for short or unrelated periods of time. Based on the assumptions of the capital asset model, beta is generally interpreted as the market contribution to the portfolio performance while alpha is supposed to measure the portfolio manager’s contribution to performance. Using dummy variables or a quadratic specification, the latter can be split further into stock picking and time picking skills. This leads us into the final stage of evaluation – performance attribution.
Attribution is the process through which we try to split the rate of return achieved into its various contributions. From a learning perspective the breakup described above between the market and the manager’s contribution is insufficient and may even be misleading. One should aim at disaggregating as much as possible to identify the contributions of each step identified in our unbundling of the investment process. For instance, it is important to distinguish the relative contributions of trade execution and closing decisions (which positions and when) to evaluate our full management skills. Likewise it is important to distinguish between the contributions of shoal selection and stock screening and those of due diligence.
Nevertheless, we immediately realize that performance attribution may be a very time consuming process. It should probably be outsourced to external consultants, unless one has a decision making platform that allows the automation of such process. As in everything else in life, investors without the necessary resources must cope by using shortcuts or more qualitative assessments, while trying to be as wide-ranging as possible.
After completing the unbundling of the main steps for successful investing we must alert the reader to the fact that these steps must not be looked upon as a single run exercise. Indeed, they have to be seen as sections of a revolving door where we always return to the start. It is important to keep track of our progress because we can learn with each run of the investment process. It is unlikely that any of us will excel in all steps of the investment process, but knowing our limitations and how far we must go to overcome them is already half way to being able to select the strategies that match our personal circumstances and guarantee us a place among the top performing investors. This is the basic idea that drove us to develop the SADIF Newsletter & Investment Management Portal; to provide investors with a set of tools organized in a coherent and disciplined way. But like in any other form of art, it is up to each individual investor to decide when to take shortcuts or to try new routes with a view to rising towards the highest levels in the art of investment.
[1] Even the so-called buy-and-hold risk free alternative referred to above is not free from more complex choices, such as the selection of which governments’ treasury bills to hold. Investors should consider the purchasing power of the net liquidation value of their financial wealth before converting it into a buy-and-hold non-financial asset. For instance consider two retiree investors, one looking forward to convert his financial wealth into a villa and golfing green fees in sunny Florida and the other wishing to convert his into tickets for traveling and cruising the world. For the first investor, the terms of trade between US Treasury Bills and Florida’s golf resorts provide the relevant benchmark. For the second, the relevant terms of trade are those between global treasury funds and the global cruising industry.
[2] Defining the correct sample size needed to achieve a given degree of closeness (e.g.15 basis points monthly with a confidence level of 95%) to the average market return is a difficult task. But, using a method of approximation we have estimated that a reasonable size needed to replicate the entire US market would require a portfolio with at least 2900 stocks.
[3] There is a wide choice in terms of how broad we want our basket. For instance we can use the Dow Jones Wilshire 5000 Total Stock Market Index that represents the stocks of nearly every publicly traded company in the three major exchanges in the United States , while globally we may use the Russell Global 10000 which covers 80 countries and all stocks with a market capitalization greater than $200 million USD.
[4] Buying the market means buying an equal share of all outstanding shares. For instance, if the market were made up of three companies issuing respectively 100, 1000 and 10,000 shares an investor wishing to buy 1% of the market would need to purchase 1 share of the first company, 10 shares of the second and 100 of the third.
[5] Including even Paul Samuelson, the most brilliant of today’s living economists.
[6] Although a top down approach is theoretically more consistent and comprehensive it may be less cost efficient. So, not surprisingly, many professionals having grown up in the industry and acquired their investment skills through learning by doing tend to prefer the bottom up approach.
[7] This might not be needed when investors wish to combine into a single portfolio stocks selected through different strategies.
|
|
| |
|
