Tuesday, September 29, 2009

Cost of Capital

This post is edited in 2016.

We are back to talking about something dry after a long, long hiatus. Ok Cost of Capital.

Basically, capital is not free, it comes at a cost.

Why is there such a cost? Well basically the person providing the capital needs to earn a return. If not, he might as well chuck it under his pillow right?

So the question is how much return does he want?

Well, the lowest return he can get without any risk of his original amt being reduced is 3% or so. That is if he buys government bonds. So the cost of capital cannot go below 3%.

Capital actually comes in two forms: debt and equity. Let's talk about the cost of debt first, bcos it's easier.

Say you want to start a company today and need money, so you go to a bank and ask for a SME loan. Depending on the nature of your business, your bargaining ability, the desperation of the loan officer, your interest on the loan should be around 6-10%, which is pretty high. Well that's bcos it's SME, may go any time one. So the bank needs some buffer. If a big Fortune 500 firm issues a bond, they can probably get US$100mn with interest rate of 4+% or so.

So the cost of debt is just that: 4% to maybe 6% for most large cap companies. In 2016, with negative interest rate dominating a lot of sovereign bonds, the long term cost of debt has come down to 2-6% for corporates.

Traditionally it has been thought that cost of equity should be higher than debt bcos the equity provider gets to participate in the upside (when the profit grows, stock price rises) but the debt owner will always only receive the fixed interest. So cost of equity will at least be 6% or more. In the 1950s or maybe 60s, academics tackled this question in a big way and came out with a huge model called the CAPM model. This is huge and Nobel prizes are given and economists became gods. For those interested, you can go wiki it or something.

Basically the idea is that cost of equity can be expressed in an equation like

Cost of equity = risk free rate + equity risk premium

The equity risk premium part can be further broken down into super complicated stuff like beta and expected market return which are too mind-boggling for our purposes here so it suffice to say that this equity risk premium should be a no. to compensate equity investors for the risk they take and make the total cost of equity higher than cost of debt.

Historically, cost of equity is about 8-10% for most large cap companies.

So the equity risk premium is about 5-7% (bcos risk free rate, which is usually long term government bond yield is about 3%)

Combining the two, you get something called the WACC (for weighted average cost of capital), and this is the cost of capital for a company. This no. usually ranges from 6-8% judging by the no.s given above.

It is said that companies should earn more than its cost of capital to justify its existence. So meaning the co. should have a return on capital of at least 6-8%. Capital meaning debt + equity, or roughly speaking total asset of the company (not exactly the same thing but close). If the co. has no debt, it means that the return on equity (the famous ROE), should jolly well be above 8-10%, ie above the cost of equity.

If a company cannot generate this return, then investors should really pull out all the funds and invest in others that can. However, in real life, that is not always the case, as we shall explore in the next post.

Monday, September 14, 2009

What's Right with Buy-and-Hold?

So Buy-and-Hold has its flaws, but the alternative, which is trading is not much better. Empirically, trading has not help generated wealth. But before we come to some conclusion, let's look at the usual Pros with regard to Buy-and-Hold.


1. Missing out the 10 biggest days in positive movement

Studies have shown that if you subtract the returns of the 10 biggest positive gain days in the past 20 yrs, your long term average annual return drops something like from 8%pa to 4%pa. This is major bcos suddenly you might as well go and buy 30 Yr Singapore Govt bonds and that could have given you close to 4%pa, without all the risks associated with stock market. Since we cannot predict which days will be the biggest positive gain days in advance, value investors argue that we should always buy and hold to reap the full benefits.

2. Transaction cost

This has come down over the years but still constitute 0.5% or more to most retail investors. Considering in Singapore where the minimal brokerage cost is S$20 per trade, you need to do a S$8,000 trade so that you can get the buy and sell costs to be 0.5% of your trade. Sadly it can only go as low as 0.25% bcos after that, the brokers charge based on the size of your trade. So let's work with 0.5%. Most retail investors would probably do more than 1 buy-sell trade per year, let's say they do 2. We know that average annual market return is 8% and two trade incur transaction costs of 1%. Congrats, you just gave your brokers a 12% commission. Every year. So Buy-and-Hold pls, save the transaction cost.

3. Dividends

Globally dividend yield usually ranges from 2-4% and in extreme times, you see dividend yield of a market going to 6-7%, like the STI in 2008. Albeit it's a lagging yield bcos the data is always late to factor in a drop in dividend going forward. Anyhow, the point is, dividend is a huge part of return. Since we all know that average market return for investment is only 8%, dividends can constitute 25-50% of this market return. It does make sense to focus a lot on dividend stocks. And needless to say, to get dividends, you can't just trade, you need to buy-and-hold.

4. Missing out the full growth

This will be the most compelling argument. Trust me.

If you have bought a Great Company, not just Good but Great with a capital G, then there is never a right time to sell bcos the company simply grows exponentially and overwhelms everything! Some of Berkshire's companies would illustrate this very well. I will just highlight See's Candy and Coke. Buffett bought over the whole of See's Candy for USD 25mn in 1972. Today See's generate pre-tax earnings of over USD 1bn, if See's is a listed co. the market cap would be close to USD 20bn. So that's close to a 1,000 fold return. Mind you, that's 100,000% return. If you sold for 100% profit in 1973, you just made the biggest mistake of your life.

Coke tells the same story. Berkshire bought 8% of Coke for USD 1bn in 1988, today the same stake is worth USD 10bn and Coke's pre-tax earnings is USD 7bn, of which Berkshire is entitled USD 600mn (on paper). In another 10 yrs, the profits entitled to Berkshire would be more than what Berkshire paid for and this current ten bagger will become a 20 bagger or even more. So is there ever a right time to sell Coke?

Sadly, most retail investors never get to enjoy this kind of thrill bcos a 20% profit in 2 weeks is already better than sex.


Again in investing, there are no hard and fast rules, most of the time buy-and-hold makes sense, but there are times that they do not, some times you can take some profit here and there during those periods. However, to trade in and out every few weeks or days and try to beat buy-and-hold is a tall order. Much taller than most people would like to think so. Not to mention the effort needed to do those weekly trades! So run through the pros and cons again, when you are tempted to trade!

Friday, September 04, 2009

What's Wrong with Buy-and-Hold?

This is another topic that has been debated left-right-centre since... Geez, Adam and Eve I guess. Let's just go through the usual pros and cons. I will start with the Cons.


1. Doesn't help to make money

This has been highlighted various times in the papers. Someone who bought an index, say the S&P500 and held for the past 10 years would have made zero return. In some markets, you could have bought-and-hold for the last 20 years and still lost money. The best example being Japan. Buy-and-hold at any point over the past 20 years would have made negative return! So to hell with Buy-and-Hold right?

2. What about locking in profits?

If a stock has risen 100%, and knowing that stocks on average gives only 8%pa, this stock has already given you roughly 10 yrs worth of return, isn't it a good idea to lock in the profits? Especially if the stock intrinsic value is not going to grow by that much over time - meaning that it's grossly overvalued. We should sell when things are grossly overvalued, right?

3. There is no time.

Buy-and-hold takes a long time to give you a good return ie average market return of 8%pa. However, time is a luxury that not everyone has, especially if you are 40 and above. You need a substantial retirement nest egg in 15 yrs assuming you retire at 55. Although retirement age is 62, most people don't get to stay employed until that age unless you are a civil servant, or self-employed. And increasingly, the career lifespan is shortening, look around you, do you see a lot of your colleagues in their 50s? So how do you buy-and-hold in 10 or 15 years? Incidentally, if you bought the STI index in Dec 99 at 2,600 points, you just managed to claw back your capital as the STI today is, well, at 2,600 points! 0%pa after 10 years, that's great isn't it?

So Buy-and-Hold sucks, what should we do?

Well you trade. You buy the STI when it was at 1,200 in Sep 2000, Hold until Sep 2007 and sell at the peak of more than 3,000. Buy back when it dropped to a low at 1,500 in Jan 2009 and sell now at 2,600. You would have make 500% gain over the past 10 yrs, that's roughly 50%pa.

Haha, well this can only be done on hindsight. Most traders lose their pants trading bcos their emotions get to them. Even if they bought at 1,200, they would have sold when it reached 2,000. Then hastily buy more when it reaches 3,000 and then got stuck when it collapsed, etc. Well you know the story, don't you.

Trading works when you have a good trading system and you adhere to it. 100%. If you do that, you can probably make an average return of 4-6%pa after transaction cost. Of course, legendary traders make a lot more than a meagre 4-6%pa. So we should strive to be like them! Then value investors will bring out Warren Buffett, who made 24%pa for 50 yrs and grew USD 1mn to USD 60 BILLON. So better argue based on average returns, not the legends' rate of returns.

In reality, most retail traders lose money trading. We are not even talking about market return of 8%pa here. Next post, we look at what's right with Buy-and-Hold.