Back in March, I wrote about the outstanding opportunities the global panic was creating in the Singapore Reit sector: It’s time to get back into property – Asian property.
At the time, fears that the credit crunch would make it impossible for many of them to refinance had hammered the sector to the point where almost every Reit sported a double-digit yield. This looked like an outstanding opportunity: the best of the bunch were very attractive income plays, with the bonus of the inflation protection that comes through owning real estate.
This turned out to be one of my better short-term calls. The sector is up an average 88% since, beating the local Straits Times Index (80%) and thrashing the FTSE (30%). I know from my emails that some of you bought into the sector at the time, and you may be wondering if it’s time to take profits. So let’s take a look and see whether there’s any value left in them …
As the panic fades, Singaporean Reits have shot up
The four Reits I singled out as especially interesting (see table below) pretty much cover the range of returns seen in the sector. How much they’ve moved has depended on how big the refinancing fears were at the time and how geared they are to the global economy. Growing confidence in a rapid rebound in Asia has been particularly good for Ascott, which offers serviced apartments for business expats and is thus seen as a recovery play.
|Ascott Residence Trust||SP:ART||S$0.88||147.50%||0.65|
|*Price rise plus dividends paid 10/03/2009 – 31/07/2009|
So after these gains, are they overvalued? To answer that, we need to think about how to value a Reit. Some investors like to look at price/book value (the price divided by the value of the Reit’s assets minus its debt). The P/B values above would suggest that A-Reit is fairly valued while the others are still under valued.
But I don’t think P/B is very useful for long-term valuation of property stocks. Remember that book value is not constant; in a property bull market it will be revised upwards, while in a bear market it will fall. Looking only at P/B is a good way to get sucked into property bubbles and hammered in busts.
What’s the best way to value a Reit?
The main attraction of Reits is the income they pay. So it makes sense to value them on their dividends (S-Reits must pass on almost all of their operating income to investors as dividends; in exchange, they pay no Singaporean tax at the company level). So we need to look at the current yields on these Reits. To do that, we need to take a couple of things into account:
1) We’ll base our estimate on the Reit’s current payout. However, we need to make sure that this is a typical, sustainable payout – in other words, we have to look through short-term factors caused by the global recession. In this case, Ascott’s first-half income is down 20% year-on-year; but this is likely to rebound pretty quickly when the economy picks up, so we’ll adjust its payout up in line with something more typical.
None of the others look likely to see exceptionally large permanent changes in revenue – positive or negative – in the near future. Suntec and, to a lesser extent, A-Reit are likely to see higher vacancy rates and lower new rents over the next couple of years, but these should recover in due course.
2) We need to allow for any other changes that could affect payouts. In Suntec’s case, unit-holders will be diluted by around 10% over the next few years as it pays for one recent acquisition in units. It also has a convertible bond due in 2013, which could mean further dilution of another 10%.
For Ascott, gearing may be a little high at 40% and it seems possible that management could carry out a rights issue to reduce it, although they say they plan to avoid one at present. At these prices, raising enough to reduce gearing to 35% would probably mean diluting shareholders by around 10%.
Taking all this into account, yields on all four still look pretty attractive compared to what else is on offer, as the table below shows.
|Name||Annualised dividend||Annualised yield||Estimated dividend||Estimated yield|
|Ascott Residence Trust||7.15S¢||8.14%||7.8S¢||8.90%|
Obviously, when assessing which offers the best value, we need to take other things into account – the overall riskiness of each Reit, the steadiness of its rents, its potential growth and so on. If you want a steady, secure income, A-Reit and Suntec are probably best. A-Reit has a large, diversified portfolio of industrial and warehouse/logistics property in Singapore. Suntec’s main asset is retail and office space in Suntec city, a prime development in Singapore’s Central Business District.
Ascott will be the most geared to global growth and the most volatile. First owns three hospitals in Indonesia, with rents pegged to both the hospitals’ revenues and inflation. That makes it a play on growth as well – since wealthier Indonesians will mean more demand for healthcare – but in a steadier way than Ascott.
A longer-term valuation method
Of course, we can try to make things more sophisticated. Since Reits are about cashflow and dividends, one way to value them is through discounted cashflow analysis. This involves looking at the future payments we can expect and discounting them back to a present value, based on the fact that a certain payment today is worth more to us than the prospect of one in a year’s time. So, for example with a discount rate of 10%, £100 now and £110 in a year have the same value to us.
However, calculating what a company is worth is a very inexact business. It involves many assumptions, and a slightly different outcome can make a big difference. Benjamin Graham, the father of value investing, advised that investors should focus on trying to identify a range of possible valuations for the stock instead.
In other words, you give up trying to be precise and settle for an approximation. Then, by using conservative assumptions and buying at or below the bottom of the range, investors can give themselves a margin of safety and hopefully avoid overvalued stocks as much as possible.
With that in mind, how do we go about calculating roughly what these Reits might be worth? Let’s start by estimating a long-term growth rate. I’m going to assume 1.5% a year for A-Reit and Suntec and 4% a year for Ascott and First.
That might sound low given that we expect Asia to grow much faster than that over the next few decades. But one of the biggest investing mistakes is to project recent growth rates off into the future – overly optimistic assumptions like that are why investors end up overpaying for growth. Very few companies can maintain double-digit growth for long periods of time. Over the long-term, most established companies seem to grow earnings at less than GDP.
We also need to choose the discount rate, which needs to reflect the riskiness and certainty of those future dividends. Using a different discount rate can make a big difference to the fair value that you come out with, but choosing them is a very subjective decision.
Personally, I look at the company’s average cost of debt – around 3.75%-4% for A-Reit, Ascott and Suntec and around 5.5%-6% at present for First. As a starting point, I double this, then make (subjective) adjustments based on how steady the company’s earnings are, how long a track record the firm has and other risks such as the threat of competition.
So, for A-Reit, with its large, diversified portfolio, I think a discount rate of 7.5% is acceptable. Suntec is less diversified and should probably require a slightly higher rate – I’ll use 8.5%.
Ascott has much greater earnings volatility and is also potentially more vulnerable to competition. Rivals can easily put up new serviced apartments in hot markets as they’re doing in Beijing and Shanghai; Ascott’s main advantage is its brand. For this reason, I’d use a higher discount rate of 11%.
Finally, First’s current cost of debt is probably overstated because of the crisis, but it has concentration risks – all three of its hospitals are leased to one company. A rate of 11% looks reasonable here as well.
A-Reit and First now look the best Singaporean Reits
So taking these numbers, we can calculate the discounted (present) value of their future stream of dividends. There’s a standard mathematical formula to do this called the constant growth dividend discount model, which is price = dividend/(discount rate – growth rate). Using it produces the results below.
|Name||Estimated dividend||Discount rate||Long-term growth||Price based on DCF|
|Ascott Residence Trust||7.8S¢||11%||4%||S$1.1|
Obviously, this is very simplistic. We could model these companies in a much more complex way – but sometimes it’s good to keep things fairly simple. Today, many analysts construct hugely complex spreadsheets of a company’s business without any apparent improvement in their ability to forecast results.
In reality, I usually take the analysis a bit further than this and play around with a few possible scenarios. For example, does it make a big difference if the recovery is very slow and we have two or three years of falling or stagnant earnings? Does it make a difference whether Suntec’s convertible bondholders convert into shares or ask to be repaid in cash?
Trying out a few of these scenarios doesn’t change the picture hugely. I get fair value estimates for A-Reit of around $2.1-$2.4, Ascott around $0.9-$1.2, Suntec around $1.1-$1.4 and First around $0.95-$1.15.
So would I sell at this point? On a long-term view, no. They’re still below my fair value estimates and I generally wait for a stock to be substantially above this before selling (I always try to minimise turnover in long-term investment portfolios, since dealing costs are a big drag on returns for private investors.) However, if I were buying in to the sector at this point, I would focus on A-Reit and First as they seem to offer the best value. For the other two, I’d hope for a correction before looking to buy in.
From a short-term perspective, I would be cautious on Ascott. It’s clearly to be treated as a global recovery play and can probably run quite a bit higher while the rally continues. However, I still expect quite a strong sell-off later this year once it becomes clear that markets are too optimistic about the speed of the rebound; in those circumstances, Ascott will be vulnerable. So if I were taking a shorter-term view, I would be inclined to take some profits on this one after such a strong run-up.
In other news this week …
|China (CSI 300)||3,735||+1.8%|
|Hong Kong (Hang Seng)||20,573||+3.0%|
|Singapore (Straits Times)||2,659||+5.0%|
|South Korea (KOSPI)||1,557||+3.6%|
|Vietnam (VN Index)||467||-2.7%|
|MSCI Asia ex-Japan||435||+3.3%|
It was an unsteady week in Chinese equities. The CSI300 index fell 5.3% on Wednesday, the biggest drop in eight months, but recovered later in the week to finish up 1.8%. The slump was triggered by fears that the excess liquidity that has helped propel the market back up could soon disappear. China Construction Bank and Commercial Bank of China, two of the ‘Big Four’ lenders, were reported to have set tight targets for loan growth in the rest of the year.
However, speculative interest remains strong for now, especially in new issues. Shares in housing contractor China State Construction Engineering rose 56% and toll road operator Sichuan Expressway soared 200% on the first day of trading.
There was more evidence that last quarter saw a strong rebound across Asia. Following strong second-quarter GDP growth in Singapore and Korea last week, the latest data from Thailand showed a solid month-on-month bounce in manufacturing and stable consumer spending after a 3.6% fall in the first quarter. Encouraging export numbers from Korea and a positive purchasing managers’ index – a survey of manufacturing conditions – from China suggest improvements continued into July.
In Japan, Honda reported a surprise profit and Nissan a smaller-than-expected loss for the three months to June. Both were helped by stronger Chinese demand; expectations for the US market remain weak.