Sunday 27 October 2013

The Best & Worst Months for Shares

In Jeremy Siegel's book "Stocks for the Long Run", it was mentioned that the best month for stocks is January while the worst month for stocks is September. October, although not the worst month, is the most volatile, having witnessed the stock market crashes of 1929, 1987 and more recently 2008. Let's review the case for Singapore to see if Singapore's stocks follow the same pattern.

The table below shows the average monthly return for the Straits Times Index based on data for the past 24 years since 1988 (2013 is excluded as there are still 2 more months to go).

Calendar Month Average Return Standard Deviation No. of
Best Months
No. of
Worst Months
No. of Best - Worst Months
Jan 0.8% 7.3% 5 2 3
Feb 0.9% 7.0% 2 1 1
Mar 0.5% 5.1% 0 1 -1
Apr 2.7% 6.5% 3 0 3
May -0.3% 7.6% 2 5 -3
Jun 0.2% 6.4% 2 1 1
Jul 1.6% 4.3% 3 1 2
Aug -2.9% 6.5% 0 5 -5
Sep -1.1% 7.1% 2 4 -2
Oct 1.0% 9.2% 1 4 -3
Nov 1.6% 5.6% 1 1 0
Dec 2.8% 4.9% 4 0 4

Based on the average monthly return, the best month is actually December at 2.8% while the worst month is August at -2.9%. However, if based on the no. of times that month has the best return in a calendar year, the best month is January with December not far behind. August remains the worst month. It is interesting to note that August was never the best month in any year. Similarly, December was never the worst month in any year. Based on the standard deviation of the monthly returns, October is the most volatile at 9.2% while July is the least volatile at 4.3%.

Hence, from the above analysis, Singapore stocks exhibit similar trends to US stocks, with December/ January being the best month while August/ September being the worst and October being the most volatile. The similarity in patterns should not be surprising, since stocks between US and Singapore are closely correlated.

From the last column of the table, it can also be seen that there is a sequence of positive returns from December to February. Similarly, there is a sequence of negative returns from August to October. Hence, there appears to be some sense in the adage "sell in May and go away".

Personally, do I follow calendar months when buying or selling shares? Actually not really. It is only a secondary consideration. The key consideration is still asset allocation and individual share performance. For example, if I intend to sell shares and January is around the corner with shares rising, I would wait till the end of January before selling. Sometimes, I might even wait till the end of February. Similarly, if I intend to buy shares and September/ October is around the corner and shares are falling, I would wait till the end of October before buying.

However, calendar effects do not always happen. Take this year for example. At end August, shares were falling due to widespread concerns that Fed might start to taper their Quantitative Easing (QE) programme. Shares dropped by 6.0% in that month. Given this backdrop, I would assume that the selling trend would continue into September and October. However, Fed did not taper QE against all expectations. Shares rose by 4.6% in September. Till yesterday, shares are up by 1.2% for October. So, calendar effects do not always happen.

In conclusion, there are certain months and quarters when shares are likely to perform better and certain months and quarters when shares are likely to perform worse. It is useful to know this information when buying or selling shares, but don't bet on it.

See related blog posts:

Sunday 20 October 2013

Do REITs Overpay for Their Acquisitions?

REITs have been active in acquiring properties to add to their portfolios. They do this to increase the rental income and distribution to shareholders as well as diversify the sources of rental income. In addition, there is also the potential for capital gains when the properties are sold. In the process of acquiring properties, they might also raise new capital from the market and increase the liquidity of the shares. However, do REITs overpay for their acquisitions in their attempts to expand? As industrial REITs are most active in acquisitions, we look at a list of acquisitions that industrial REITs have entered into with other companies.

Below is a table listing the acquisitions industrial REITs have carried out with other companies listed on the SGX, showing the acquisition price, valuation price and premium above the valuation price.

Property Date Price Buyer's Valuation Price - Valuation % Premium
65 Ubi Ave 1 May 2003 $35.0 $35.0 - -
1 Changi Business Park Ave 1 Sep 2003 $18.0 $18.0 - -
TT Int Tradepark Nov 2003 $92.0 $92.0 - -
12 Woodlands Loop May 2004 $24.8 $24.8 - -
9 Changi South St 3 Oct 2004 $32.0 $32.0 - -
5 Toh Guan Rd East Oct 2004 $36.4 $36.4 - -
52 Serangoon North Ave 4 Feb 2005 $14.0 $14.0 - -
84 Genting Lane Jul 2005 $10.0 $10.0 - -
20 Old Toh Tuck Rd Nov 2005 $11.6 $11.6 - -
50 Kallang Ave Dec 2005 $28.6 $28.6 - -
2 Serangoon North Ave 5 Dec 2005 $45.0 $45.2 -$0.2 0%
31 Ubi Rd 1 Dec 2005 $23.0 $23.0 - -
2 Senoko South Rd
26 Senoko Way
Oct 2006 $49.0 $49.0 - -
30 Woodlands Loop Nov 2006 $10.3 $10.4 -$0.1 -1%
20 Tampines St 92 Dec 2006 $10.0 $11.2 -$1.2 -11%
134 Joo Seng Rd Dec 2006 $10.7 $10.9 -$0.2 -2%
3 Changi South Lane Dec 2006 $13.9 $14.4 -$0.5 -3%
521 Bukit Batok St 23 Dec 2006 $24.1 $24.9 -$0.8 -3%
9 Tampines St 92 Dec 2006 $11.0 $11.0 - -
31/33 Pioneer Rd North
119 Neythal Rd
30 Tuas Ave 8
8 Tuas View Square
Aug 2007 $36.8 $37.6 -$0.8 -2%
31 Int Business Park May 2008 $246.8 $246.8 - -
1/2 Changi North St 2 Aug 2010 $22.1 $22.2 -$0.1 0%
25 Tai Seng Ave Sep 2010 $21.1 $21.5 -$0.4 -2%
4/6 Clementi Loop Mar 2011 $40.0 $40.0 - -
3C Toh Guan Rd East Nov 2011 $35.5 $35.5 - -
16 Tai Seng St Mar 2012 $59.3 $59.3 - -
15 Jurong Port Rd Dec 2012 $43.0 $43.0 - -
$37.2 $37.3

From the table above, we can see that most of the transactions are carried out at the valuation price. A minority of the transactions are at a slight discount, ranging from 1% to 11%. The average discount of all transactions is 1%. On this basis, REITs do not overpay for their acquisitions.

As the counter-parties to these transactions are also listed companies, the sellers might also announce the transactions on SGX. Some of the sellers would disclose their valuations of the properties, thus allowing us to see both sides of the transactions. The list of transactions with sellers' valuations is shown in the table below.

Property Date Price Seller's Valuation Price - Valuation % Premium
65 Ubi Ave 1 May 2003 $35.0 $35.0 - -
TT Int Tradepark Nov 2003 $92.0 $92.0 - -
20 Old Toh Tuck Rd Nov 2005 $11.6 $9.3 $2.3 25%
31 Ubi Rd 1 Dec 2005 $23.0 $23.0 - -
28 Senoko Dr Apr 2007 $12.0 $12.0 - -
31/33 Pioneer Rd North
119 Neythal Rd
30 Tuas Ave 8
8 Tuas View Square
Aug 2007 $36.8 $22.6 $14.2 63%
1 Kallang Way 2A Jan 2008 $14.0 $12.0 $2.0 17%
31 Int Business Park May 2008 $246.8 $246.5 $0.3 0%
29 Tai Seng Ave May 2010 $53.0 $40.0 $13.0 33%
1/2 Changi North St 2 Aug 2010 $22.1 $14.3 $7.8 55%
25 Tai Seng Ave Sep 2010 $21.1 $16.6 $4.5 27%
44 & 46 Changi South Rd Dec 2010 $16.8 $12.5 $4.3 34%
4/6 Clementi Loop Mar 2011 $40.0 $22.0 $18.0 82%
3C Toh Guan Rd East Nov 2011 $35.5 $31.0 $4.5 15%
16 Tai Seng St Mar 2012 $59.3 $59.0 $0.2 0%
15 Jurong Port Rd Dec 2012 $43.0 $33.0 $10.0 30%
$47.6 $42.6

From the table above, it is interesting to see that most of the transactions are carried out above the seller's valuation price, with premiums ranging from 0% to 82%. The average premium of all transactions is 24%. Note that these valuation prices are not historical costs at book value; they are obtained from recent valuations commissioned by the sellers.

So, we are back to the original question: do REITs overpay for their acquisitions? To answer this question, we need to address 2 other questions, i.e. why do buyer's and seller's valuations differ, and the relative bargaining powers of buyers and sellers.

It is not uncommon for valuations of the same property to differ. Based on the disclosure in the SGX announcements by both buyers and sellers, 2 key reasons could be distilled:
  • Buyer's valuations are more rigorous than seller's valuations. Buyer's valuations are generally based on 3 or more valuation approaches, some of which include direct comparison of recent transactions, discounted cashflow of income stream, capitalisation of income stream and replacement cost of the property. In contrast, seller's valuations (note: there are only 5 sellers that disclosed the valuation approaches used) are based on less approaches, including desktop valuation. Furthermore, some of the acquisitions need to be funded by bank loans. If the properties are overvalued, banks would run the risk of loss. On this basis, the "correct" value of the property probably lies closer to the buyer's valuation.
  • Basis used in valuations. The basis used in valuations play a key role in the eventual value of the property. In one of the sellers' announcements, it mentioned about valuations on a sale-and-leaseback basis and on a vacant possession basis. Needless to say, the valuation on a sale-and-leaseback basis should be higher than on a vacant possession basis given that the leaseback agreement provides a rental guarantee for several years.

On the relative bargaining powers of buyers and sellers, even though the seller might have a valuation price, it does not necessarily mean that the seller would be willing to sell at that price. The seller probably had been operating from the property for several years already and enjoying continual appreciation in property values. Some premium above the seller's valuation price is required to entice the seller to sell the property and give up future capital gains as well as converting from being a property owner to a property lessee.

Lastly, of all the properties mentioned above with the latest valuation, only 2 properties have seen their valuations fall below the original purchase price.

In conclusion, on the basis that (1) the acquisition price does not differ much from the buyer's valuation price, (2) the "correct" value of the property probably lies closer to the buyer's valuation than the seller's valuation, (3) a premium is likely required to entice the seller to sell the property, and (4) the current valuation of most properties mentioned above is equal to or above the original purchase price, REITs do not overpay for their acquisitions.

See related blog posts:

Sunday 13 October 2013

Managing Expenditure

Saving is usually the first step towards achieving your financial objectives. One approach of building up savings is to have additional sources of income. The other approach is to defer consumption to a later date. To manage my expenditure, I have tried several ways over the last 2 decades; some work and some don't quite work. Let's discuss the ways that don't quite work before discussing the ways that work.

Ways That Don't Quite Work

The conventional way to manage expenditure is to set a budget for the various categories of expenditure. After trying for several years, I have come to the conclusion that budgets don't work. It works well for expenditure that are regular, such as utilities, taxes, etc. But when it comes to discretionary expenditure such as recreation, social events, etc., the budget is sometimes exceeded. The main reason is that the budget does not come to mind at the time of spending money. By the time I remember about the budget, money is already spent.

Related to setting a budget is to record all the expenses made. By recording and compiling all the expenses, you can analyse at the end of the day/ week/ month whether the money has been spent wisely and if not, to remind yourself to spend wisely the next time. Again, it does not quite work. Regret only lasts for as long as I'm looking at the expenses. It does not last until the time of spending money.

Perhaps the only time that budget setting and expense recording work is near the end of the month. If the budget is almost exceeded, I would try to defer some expenses to the next month.Conversely, if I spend less in this month, I would bring forward some expenses from the next month so that more money is available for spending next month. This is usually done by topping some stored value cards that are regularly used.

Ways That Work

The ways that work in managing expenditure (at least for me) has to do with how we think about expenditure. There are 2 ways of thinking about expenditure. The first way is to think of it in terms of opportunity cost. When you spend money on certain items, it means that you can't spend on some other items. As investors, it also means that the money cannot be invested and grown. Hence, the opportunity cost to investors is usually very high. The higher the rate of return and the younger the investor is, the higher is the opportunity cost. As an illustration, I have a target rate of return of 10% per year. At this rate of return, money will double every 7.2 years. I have 27 years before reaching the retirement age of 65. So, every $1 today will become $14.81, say, $15, when I retire. That means that for every $1 I spend today, I will have $15 less when I retire. Hence, the price tag of every item today is multiplied by 15. A mid-level digital camera that costs $2,000 to most people will actually cost $30,000 to me. That is a lot of money! From this perspective, I would think very hard about spending money on big-ticket items.

The second way of thinking about expenditure is to think about the amount of hours you need to put in at work to pay for the item. Let's say the expenditure is equivalent to 2 weeks of salary. Instead of the company paying you a monthly salary and using part of the salary to pay for the item, would you be willing to work for FREE for 2 weeks and at the end of the 2 weeks, the company gives you the item? Bear in mind that during these 2 weeks, you will need to endure the usual scolding from the bosses, tight deadlines, office politics, etc. Very likely, you will think a little longer about spending the money.


Generally, my experience is that if you manage the big-ticket expenditures well, you do not really need to worry too much about the small-ticket expenditures. You might overspend on the small-ticket items, but they will not blow a big hole in your pocket. Managing expenditure is not about being a miser; it is about knowing when to spend.

See related blog posts:

Sunday 6 October 2013

Clean Out CPF Balance When Taking HDB Housing Loan?

When you take a housing loan from HDB, it will take all the money in your CPF Ordinary Account (OA) before giving you the loan. This is to reduce the loan quantum that you need to service. To avoid having an empty OA account, you'll need to transfer some or all of the OA balance to somewhere else before you apply for the loan. One way is to invest in some bond or money-market unit trusts that will not drop too much in value while you're holding them. But the key question is, do you want to leave some money in your OA account or clean it out? It depends on whether you have any need for the OA money in the future and your risk preference.

There actually isn't a lot of approved uses for the OA money. Besides building up your retirement nest egg, you can buy properties, service insurance policies, invest in stocks and unit trusts and finance your family members' tertiary education. If you foresee that you need to use the OA money for such needs in the short- and long-term, then it's better to keep some money in the OA. This is because if you have an empty OA account, it'll take a fairly long period of time to build it up again as the monthly loan repayment will take a portion of your monthly CPF contributions. In any case, assuming you keep some money in the OA but later find that you have no other need for it, you can always make a lump sum repayment to reduce the loan quantum.

The second factor that you need to consider is your risk preference. Do you prefer a bigger debt burden (and bigger OA balance) or a near-empty OA balance (but smaller debt burden)? Some people will prefer to have a smaller debt burden so as to reduce the amount of interest payable over the loan tenure and/or to become debt-free earlier. Personally, I prefer to have a bigger OA balance. The main reason is it takes a long time to build up the OA balance when you start to service the loan as explained earlier. If, for some reasons, you are in between jobs for several months, the OA balance will quickly become insufficient to service the monthly repayment and you might risk losing your flat (I understand HDB is quite forgiving if you miss a few monthly payments, but the risk exists). Not only that, you might also default on other financial commitments such as insurance premiums, risking your insurance coverage as well. For this reason, I prefer to live with a bigger debt burden than to risk a snowball default. Of course, you can always make up any shortfall with cash, but it would be a constant worry over the low OA balance.

It also helps that the HDB loan interest is only 0.1% higher than the OA interest rate. At 0.1%, you pay an extra $100 for every $100,000 in your OA balance that is not being used to repay the loan. Over a 25-year loan tenure, this works out to be $3,112. This extra payment is like insurance premium, to guard against a snowball default and constant worry over the low OA balance. Not only that, you get to save the money for some other uses in the future and have the flexibility of making a lump sum repayment. And if you manage to invest the OA money at a rate higher than the loan interest, you get to grow a bigger retirement nest egg as well!

See related blog posts: