Saturday, 24 August 2013

The "Prudent Family" Rule in Portfolio Management

In portfolio management, there is a rule known as the "Prudent Man" rule. This rule dictates that an investment professional who is entrusted to invest a trust fund on his clients' behalf has to act as what a prudent man would have done if there were no specific investment instructions. This post is not about the "Prudent Man" rule, but is a light-hearted post introducing a new "Prudent" rule, which is the "Prudent Family" rule. Note that although the Prudent Family also has a Prudent Man as a family member, he is not the same man as in the "Prudent Man" rule. For the rest of this post, "Prudent Man" refers to the Prudent Man in the Prudent Family, not the Prudent Man in the "Prudent Man" rule. 

The "Prudent Family" rule provides a light-hearted way of looking at portfolio allocation among various asset classes. By following the "Prudent Family" rule, you will find that you will end up with a balanced portfolio of stocks, bonds, hybrid securities/ alternative investment and cash. The largest proportion will be stocks, followed by bonds, hybrids/ alternatives and cash. Let me now introduce the various members of the Prudent Family.

Firstly, who in the Prudent Family would find it prudent to hold stocks, which offer high growth but are  more volatile? Needlessly to say, it would be the Prudent Man, who will want to have some adrenaline and capital appreciation in the Prudent Family Portfolio. Also, being the head of the household, he will have the largest say in the portfolio. Following stocks in the Prudent Family Portfolio will be bonds, which are almost the direct opposite of stocks with lower growth but higher stability. So, who in the Prudent Family would think it prudent to hold an opposing view from that of the Prudent Man? That would be the Prudent Woman. As the saying goes, behind every successful man is a woman. Being the woman behind the Prudent Man, she will have the second largest say in the family portfolio.

Next on the Prudent Family Portfolio are hybrids/ alternatives. Hybrid securities, as the name suggest, have characteristics similar to both stocks and bonds. Examples would be convertible bonds, which pay regular coupons (i.e. dividends) like normal bonds but can rise and fall in value with a corresponding rise or fall in the mother shares. Even REITs can be considered to fall into this category. Alternative investments, on the other hand, bear little or no relationship to the traditional asset classes of stocks and bonds. Examples would be commodities, which do not pay regular dividends and are most of the time not correlated to the price movement of traditional asset classes. Coming back to our Prudent Family, who would think it prudent to have characteristics of both the Prudent Man and the Prudent Woman or be totally independent of them (i.e. do not take any sides)? That would be the Prudent Child. Being the youngest member in the family, the Prudent Child has a small say in the family portfolio, especially if he prefers full independence from the Prudent Man and Prudent Woman.

Finally, the last on the Prudent Family Portfolio is cash. As we all know, cash earns very little in interest and is easily eroded by inflation. Given such characteristics, who in the right mind in the Prudent Family would think it prudent to hold cash? That must be the Prudent Mistress, isn't it?

So, in the Prudent Family, the Prudent Man would hold stocks for growth, the Prudent Woman would hedge against the Prudent Man with bonds, the Prudent Child would either hold hybrids or alternatives so as not to take sides with either the Prudent Man or the Prudent Woman, and the Prudent Mistress would embrace cash for safety.

It is worth highlighting that for the Prudent Family Portfolio to work well, every member of the Prudent Family has to be prudent. For example, if the Prudent Mistress becomes imprudent and holds too little cash, the entire portfolio will be subject to the depressions of Mr Market which the Prudent Family cannot fight back against. Conversely, if the Prudent Mistress becomes too prudent and holds too much cash (which means that the rest of the Prudent Family become imprudent), the portfolio will be eroded by inflation. You can work out the outcomes on the portfolio if the other members of the Prudent Family become imprudent. In all cases, the outcomes will be negative. Hence, every member of the Prudent Family has to play his/her respective role well to ensure that the Prudent Family Portfolio can achieve its investment objectives.

If you find the Prudent Family is an interesting family, there's still one more person in the Prudent Family for you to meet. That would be its originator, (The) Boring Investor :)

Sunday, 18 August 2013

The Hidden Risks of Buy-and-Leasebacks for Industrial REITs

Buy-and-leaseback arrangements are quite common among some of the industrial REITs. They buy a property and lease it back to the original seller for a no. of years. The advantages of this practice are it increases the distribution to shareholders, diversifies the sources of rental income and has the potential for capital gains when the property is eventually sold. However, there are risks involved in this practice.

While the property is a long-term asset often lasting 50 years or more, the lease agreement is a much shorter one that lasts between 3 to 7 years (with extensions at the options of the tenant). The thinking behind buy-and-leaseback arrangements is that the REIT would be able to find a tenant and maintain or increase the rental income for the intended holding period of the property. However, some of the industrial properties are highly customised ones. The pool of potential tenants would likely be limited to those in the same industry as the original tenant (Of course, there is always the option of pulling down the property and redeveloping it). Given the small pool of potential tenants, would there be sufficient competition for the industrial space to maintain or increase the rental income? This has not even considered the fact that Singapore is reducing the intake of foreign labour that many industries depend on and there are less expensive places overseas to do business. If the rental income is reduced, the value of the property will take a hit. Yet, the loan originally taken to finance the acquisition remains unaffected and still needs to be serviced and eventually paid off. There will be lower distribution for shareholders. In addition, with a lower valuation, the debt-to-asset headroom will be reduced, leading to less property acquisition opportunities in future and moving the REIT closer to the Loan-To-Value (LTV) covenants in its debt obligations. Of course, the extent of the impact will depend on the size of that particular property relative to that of the REIT's property portfolio.

Conservatively, the buy-and-leaseback arrangement should be treated as a financing transaction for the seller with the property as the collateral. However, in this respect, REITs are at a distinct disadvantage compared to banks. Let us compare the case of a company taking a loan from the banks using the property as the collateral versus the case of the company engaging in a buy-and-leaseback arrangement with the REITs. Banks will have the following advantages over REITs:
  • Banks can request for other collaterals, whereas the only collateral for REITs is the property.
  • Banks can give a lower loan amount than the value of the collateral, whereas REITs usually pay close to the valuation of the property.
  • Banks can monitor the valuation of the collateral and ask for margin top-ups if the LTV ratio rise above a certain threshold. REITs will have to absorb the reduction in valuation.
  • Banks can demand immediate repayment of the loans and seize the property for sale if the company does not service its loans. REITs can only fall back on the rental security deposits.
  • Bank loans usually contain cross-default covenants, i.e. if the company defaults on one loan, it will trigger automatic defaults on all other loans and banks can demand immediate repayment of all these loans. Lease agreements do not have such covenants and cannot ride on the cross-default covenants of bank loans, i.e.a default on a lease agreement does not trigger a default on the loans.

It is worth noting that if the REIT takes on a loan to finance the buy-and-leaseback transaction, it is actually on the short side of both transactions, i.e. it is subject to the protective features of the loan transaction but the lease transaction do not offer comparable protection.

In conclusion, a buy-and-leaseback arrangement provides opportunities for REITs to increase their rental income and distribution to shareholders. However, there are also risks involved. REITs will need to exercise caution before entering into such transactions.

See related blog posts:

Friday, 9 August 2013

A Comparison of Risk Management Tools & Strategies

In my last 2 posts, I have discussed some of the risk management tools and strategies that can be used to manage investment risks. This post puts together all the risk management tools and strategies and discusses their pros and cons.

Tool #1: Value-at-Risk Analysis

Value-at-Risk (VaR) analysis was discussed at length in the post on Stress Testing Your Portfolio. Essentially, this tool serves to estimate the amount of loss at a given probability of occurrence and in a given holding period based on historical price changes. The advantages of this tool are:
  • It provides a sense of the loss expected for a decline. By estimating the amount of loss in advance, an investor is better able to prepare himself psychologically for the decline and make rational decisions at the depth of the bear market.
  • It allows adjustments in the portfolio to be made. If the estimated loss is too much to bear, an investor can adjust his portfolio to something more palatable.
  • It identifies the stocks that are likely to decline the most. Based on this information, an investor can determine if he should still keep those stocks in his portfolio.

However, VaR analysis is not fool-proof. The estimated amount of loss from VaR analysis may or may not be the same as that actually experienced, as the actual decline may have a different probability of occurrence or duration of decline. The disadvantages of this tool are:
  • It relies heavily on historical price changes. When the financial conditions are different, the historical price changes may not mirror the actual price changes. For example, the Global Financial Crisis (GFC) was of a different nature from the Dot.Com Bust. Using price changes from the Dot.Com Bust to estimate the VaR losses during the GFC might underestimate the actual losses.
  • New stocks have limited price history. Similarly, VaR estimates for new stocks with limited price history might be understated.
  • Duration of a market decline is seldom equal to a year. VaR is usually estimated for a month or a year. However, a market decline is seldom one year long. The longer the duration, the higher is the loss.
  • Detached from real market events. Although VaR analysis uses historical price changes, it does not reflect a real event, such as the Dot.Com Bust or GFC. The VaR estimates feel like a "statistics" and not potential losses from a real market decline.

Tool #2: Back-Testing Using Actual Market Events

Back-testing using actual market declines corrects for the detachment of VaR analysis from a real market event. It achieves similar objectives as VaR analysis, which is to estimate the potential losses from a market decline. It works even simpler than VaR. To perform this analysis, just take a real market decline, say, the GFC, measure the total decline from the peak to the trough, and apply to the exposure you have to the stock. For example, if you have a $10,000 exposure to a stock, and it declined 40% during the GFC, the estimated loss from this stock is $4,000. Repeat this for all stocks in the portfolio and you will get the estimated loss for the portfolio. As back-testing also relies heavily on historical price changes, it shares many of the pros and cons of VaR analysis. Specific advantages of back-testing are:
  • The loss estimate is based on a real market event. It is able to answer the question, "If your portfolio were to go through the GFC again, what kind of losses would it have?". By linking it with a real market event, investors are better able to relate to and prepare themselves for the potential losses.
  • It provides a realistic check on the estimated losses from VaR analysis. Unlike VaR analysis, the estimated losses from back-testing is not just a "statistics". It also corrects for the duration that affects VaR estimates.

Specific disadvantages of back-testing vis-a-vis VaR analysis are:
  • It does not show the correlation between different stocks in the portfolio. Back-testing only consider the peak-to-trough decline and does not consider the price changes in-between. Hence, the correlation between different stocks cannot be performed for a deeper analysis.

Nevertheless, VaR analysis and back-testing based on actual market events are not mutually exclusive. They can work hand-in-hand to serve as a cross-check on each other.

Strategy #1: Diversification

Everyone should be familiar with this strategy, so I will not discuss too much about it. Just take note that during market crises like the GFC, most types of investment will go down together.

To share my experience during the GFC, I had a fairly diversified portfolio at the start of the GFC. The table below shows the portfolio allocation among different asset classes in Jan 08. Also shown is the % change in value (ignoring dividends) in Oct 08, at the depth of the crisis. (Note: The % change assumed that the portfolio remained constant throughout the period).

Shares REITs Biz Trusts Preference Shares SGS Bonds Cash Total
% Allocation 18% 7% 11% 18% 12% 33% 100%
% Change -64% -66% -65% -22% 3% 0% -27%

As shown in the table, diversification worked to some extent, but not as well as expected prior to the crisis. Both REITs and Business Trusts (BTs) fell as much as shares and served no diversification benefits. Preference shares fell by a less extent, but they were expected to hold steady. Only the Singapore Government Securities (SGS) bonds and cash held steady as expected.

Strategy #2: Never Run Out of Cash

Cash is ammunition. When market prices are severely depressed, having cash on hand allows you to buy on the cheap. Without cash, you will be at the mercy of the market. Psychologically, cash provides a form of security and allows you to be in control. Having said that, buying further might lead to further losses, but at least you have a choice and are not dependent on the market or on others to bail you out.

Although cash is a great friend in times of crisis, holding too much cash might erode their value in times of high inflation like the past 5 years. So, holding them in cash-like instruments such as preference shares and SGS bonds would help to minimise inflation losses. See the post on Behind Every Successful Bear Market Recovery is A Cash-Like Instrument for the experience with preference shares.


The above are some of the risk management tools and strategies that can be useful in managing investment risks. They work to some extent, but they have some shortcomings as well.

See related blog posts:

Saturday, 3 August 2013

Behind Every Successful Bear Market Recovery is A Cash-Like Instrument

29 July 2013 marks the end of the OCBC 5.1% Non-Cumulative Preference Share (NCPS). I had held this stock steadfastly since its inception 5 years ago, only liquidating it partially to raise cash for the equities and REITs during the Global Financial Crisis (GFC) in Oct 08. I seldom think about this stock, but now that it is gone, I'm beginning to remember its importance to my portfolio and investment strategy. This blog post is dedicated to this stock which has served me so well in the past 5 years.

Firstly, to understand its importance to my investment strategy, you have to understand that I have an equity-centric investment strategy. This means that I think of asset allocation primarily as between equities and non-equities. As to what the non-equities should comprise is a second-order question, so long as I can liquidate the non-equities when the equities portion calls for it (see the post on Have a Plan for a brief description of this investment strategy). Ironically, the most important part of this equity-centric investment strategy is not the equities portion, but the non-equities portion. When there is a bear market, most shares will decline, almost independent of which shares you bought. At this time, you'll wish to have a cash pile to invest in equities at now much lower prices. The ability to raise cash from the non-equities portion then becomes key, without which you are basically at the mercy of the market. To be a good cash-like instrument, it must be able to raise cash quickly and at minimal loss. If cash cannot be raised when needed or at an unacceptably large loss, it would jeopardise the opportunity to buy low and sell high subsequently. While cash is the most obvious choice, it suffers from the ravages of inflation, which for the last 5 years, was at an unusually high level. On the other hand, OCBC 5.1% NCPS (and its elder sibling, OCBC 4.5% NCPS, before its inception) paid a 5.1% dividend semi-annually, which was just able to keep up with inflation that averaged 4% a year from 2008 till 2012. It was also a very stable stock, exhibiting very little volatility and correlation with the stock market. Its biggest one-day decline was around 2.5% every 6 months, on the days it went ex-dividend. So, OCBC 5.1% NCPS became the key cash-like instrument of the non-equities portion.

Was it able to perform its role as a cash reservoir for the equities portion when the time called for it? There was only one time in the last 5 years when it was called into action, which was during the GFC. Unexpectedly, due to the financial nature of the GFC, it fell by as much as 17% at the depth of the crisis in Oct 08. I sold half my holdings at 8% loss in that month to invest in REITs and shares, which fell by even more. So, despite losing money in selling it, I more than made up for it with the subsequent rise in REITs. In the one instance when it was called for, it has performed its role admirably, even though it did not quite fulfil the criterion of minimal loss.

While the NCPS was the non-equities of choice, it was not the only non-equities instrument I had. Prior to the GFC, I had presumed that REITs and Business Trusts (BTs) would only record small losses during a bear market and could still be liquidated to support the equities. To my horror, the REITs and BTs fell by as much as the equities. Most of the BTs never recovered and ended up as major losses. As for the REITs, they had fallen so much that it did not make sense to liquidate them. In fact, they were competing for cash with almost 1-for-1 rights issues at near 40% discounts to the prevailing market prices (which had already fallen off steeply)! You never knew which ones in your portfolio would be the next ones offering a rights issue. So, you had to reserve some cash to cater for the REITs' rights issues at a time when the equities portion were also screaming for cash injections. From this perspective, OCBC 5.1% NCPS performed really well among the non-equities as a cash reservoir.

The second key role the NCPS played was in preventing a major loss in other high-yield instruments. Around the time of its inception, the minibonds were quite popular with their handsome dividends. I was also tempted by their dividends and the role they could play as a cash reservoir. However, considering that firstly, the NCPS' dividend was comparable to that of the minibonds; secondly, the NCPS' share price could be observed directly; and thirdly, the NCPS could be sold and settled in 3 working days, the idea to invest in the minibonds was abandoned.

Was there any other areas that the OCBC 5.1% NCPS contributed? It and its 4.5% sibling showed that a market pricing discrepancy could continue for a very long time. In the post on Preference Shares, I had explained that because of the 12% difference in their dividends, the 5.1% NCPS should be priced about 12% higher than its 4.5% sibling. But the market only priced in a 4% difference (after accounting for a difference in the timing of the dividends). Throughout the 5 years of its existence, this discrepancy wasn't corrected.

To conclude, OCBC 5.1% NCPS was truly boring stuff. But it saved my portfolio and allowed my equity-centric investment strategy to withstand the market test of a lifetime. Thank you so much. If there were any regrets, it would be that I had seldom appreciated its importance and that it is no longer around.

See related blog posts: