UBS Warburg has spent the last three years lobbying the government and trying to launch a real estate investment trust (REIT) market in Singapore. Yesterday's decision to pull the IPO for the Lion State's maiden S-REIT has thus left people at the firm uncharacteristically embarrassed and "shellshocked".
The deal, which was also led by DBS, was set to open up one of the most exciting investment banking opportunities in Asia. Copying the success UBS Warburg has had had in the $20 billion Australian REIT market, the CapitaLand SingMall deal was to be a beacon for other issuers such as Keppel Land and Centrepoint Properties. It was also to offer Singapore banks an avenue for disposing of their property portfolios, a thing they must do by 2004, according to the MAS.
However, yesterday's S$740 million IPO for SingMall had to be pulled only hours before the retail tranche was due to close, with an announcement sent out at around 9am.
What went wrong?
Quite a lot it seems, with some rival banks claiming the deal was totally mispriced.
First the facts. UBS Warburg and DBS failed to persuade the average Singaporean of the merits of the S-REIT product. In spite of a widespread advertising campaign, the leads garnered less than S$50 million of demand through retail brokers and DBS's ATMs network. Only 12% of the demand came from retail and the total offer was 20% undersubscribed.
A UBS Warburg official says the decision was taken to pull the deal on Sunday night, because they knew any deal would tank in the aftermarket and poison the S-REIT market forever after.
Both UBSW and DBS had hoped to lure retail investors to buy a large chunk of the offering thanks to what they were marketing as its defensive quality. The S-REIT offered a yield of 5.75% and thus gave a substantial pick up to the 3.3% 10 year government bond and the 2.1% paid on a 12 month time deposit.
A REIT is a pure property play, where investors buy a trust which holds various buildings. Investors are paid a yield each year based on rental incomes. Meanwhile no earnings are retained by the trust. The buildings in the trust are revalued every year, and it is thus the most transparent property vehicle an investor can own, with investors having access to every detail of leases and their expiry dates. Unlike a listed property company, it doesn't undertake any development activities.
However, the allegedly utility-like characteristics of the deal either failed to be understood by the average Singaporean or were considered unrealistic. While the bond-like quality of the product should have had appeal, the fact that the capital was not guaranteed caused concern. There was also a strong suspicion that with Singapore in its worst recession in years, the underlying rents on which the yield is based would also fall off.
This fear was partly as a result of the assets that had been injected into the REIT. These were the Tampines shopping mall, Junction 8 and the Funan IT Mall.
Those who doubted the S-REIT's defensive quality did so because they worried that Singaporeans would curtail their shopping and this would lead to vacancies in the malls and thus lower rents. Lower rents would mean the REIT would be serviced with lower yields.
The deal also came in a week when DBS was boosting its own capital base via a S$2.1 billion equity offering and scooping demand out of the market and using its own ATM network for this task too.
Meanwhile demand from domestic institutions such as insurance companies was lukewarm. The most sophisticated European real estate buyers, the Dutch, were not big buyers and thanks to the fact that CapitaLand didn't want to open itself up to potential litigation in the US for its forward-looking statements, the deal (amazingly) did not have a 144a. It thus could not be sold to US investors, the biggest REIT buyers in the world, with an $80 billion market.
Asian funds were not very excited either. After deducting withholding tax, the yield fell to 4.5%, and paled in comparison to Australian REITs offering an 8% yield.
Rival investment banks were recommending that their institutional clients did not buy the deal. CLSA told its clients to avoid the stock, and Salomon Smith Barney put out a research note in which it told its clients it saw "limited upside" for the deal.
"We believe retail rents are likely to be adversely affected by the economic downturn," said the report. "Hence it is prudent to assume at least a 5% decline in market rents for 2002."
It had issue with the fact that the leads reckoned that tenancies expiring in the remainder of 2001 and 2002 would be renewed at the same rates as in October. And that tenancies expiring in 2003 could be renewed at a rate 3% higher.
Meanwhile, since the malls were all 100% occupied, "going forward it will be harder for management to grow rental income", thus capping the upside.
The major issue that Salomon and others picked up on was something that the leads had been remarkably silent about. Marketing the yield story aggressively, the lead managers chose not to say very much about the underlying net asset values of the properties being bought.
The differences of opinion about the valuations are significant, to say the least. Salomon reckons that Tampines Mall was injected at a "significant premium" and the overall premium of the properties to net asset value was 15%.
CLSA valued the S$895 million of properties at a lower level too, reckoning they were overvalued by 8.3%.
UBS Warburg says that the properties were injected at neither a premium nor a discount to net asset value.
This is very significant, since almost everyone thinks the net asset values that Singapore property companies publish are too rich. This is reflected in the way the market values them. Singapore Land, for example, trades at a 43% discount to its net asset value.
The net asset values have to be determined, by law, every three years and professional valuers such as CB Richard Ellis and Jones Lang Wootton do this job. Obviously, the equity market does not necessarily agree with their valuations.
Valuing properties is tricky when there are few benchmarks. However, when actual buildings get sold that offers concrete proof of where land values lie. The most recent sale was by British insurer CGU which sold its 12 storey building in the financial district at a 23% discount to its net asset value last month.
ING Barings forecasts that prime office space capital values are set to decline from S$1,436 per square foot to S$1,280.
The broad view is that all property in Singapore is similarly overvalued. This means that by injecting the shopping malls into the trust either at net asset value or a slight premium to it, probably substantially overvalues the properties in question. No matter how good the yield of the REIT, this is a substantial problem, since the underlying value of the stock is more dependent on the value of the buildings.
According to rival bankers, arbitrageurs were waiting in the wings for when the deal was launched to short SingMall.
The logic goes that if the properties were being injected at an inflated price, the stock price would have to come down to reflect this. This exact same principle occurred with Nomura's recent J-REIT for Office Building Fund of Japan, which has seen it stock price sag since its launch in September, in recognition of the fact the market thinks the underlying properties were overvalued. Conversely, Nikko Salomon Smith Barney's similarly-timed deal for J REA has held rock steady, based on a more realistic valuation of the buildings within the trust.
Rival banks think institutional investors baulked at the SingMall valuation, and that the hope was that less sophisticated retail investors would not know what an NAV was. In the event, they refused to buy too.
The whole affair is a massive blow to UBS Warburg, DBS and Singapore Inc, all of whom staked a lot on getting this deal done.
In the case of UBS Warburg, it relocated its premier real estate expert, Andrew Pridham and a team of six from Sydney to foster the S-REIT market and try to dominate it, as it does the more mature Australian market. As a further mark of it seriousness it held a REITs conference in Singapore for 300 investors earlier this year.
UBS Warburg was the driving force behind lobbying the government to change various tax laws to make S-REITs possible. For example, the key issue of tax transparency was recently introduced by Singapore, meaning that investors would not be taxed at source on their dividends. Persuading the Singapore tax authorities to make this change had been one of the key planks in UBS Warburg's mission to make REITs possible.
This was not an entirely altruistic mission. Big fees beckoned. The firm reckoned that S$43 billion of investment grade office and retail property existed in Singapore and that S$13 billion would be injected into REITs by mid-2004. That would make it even bigger than the existing listed property sector.
The importance of REITs for Singapore is fundamental, as it offers a way out of a very key impasse the economy faces. As an ING Barings report states in May: "We believe the establishment of S-REITs is vital to break the current restructuring impasse. The strategies of many listed entities, including the banks, are precariously pegged to the divestment of property assets, but there are simply not enough ready buyers to absorb the supply glut. Our conviction is premised on the need to stimulate capital liquidity flows to the investment market by creating new demand through pooling arrangements. Otherwise, corporate Singapore risks a prolonged shrinkage in its overall market capitalization if the deadlock is not resolved."
In the case of the banks, they own 26% of Singaporean property. They need to liquidate this by 2004, and use the proceeds to become regional financial players.
Meanwhile the property companies, like CapitaLand, want to emulate the likes of Westfield Holdings in Australia which thanks to REITs have been able separate property development from property ownership. Westfield has gained by taking a management fee for looking after the properties in the trust and raising capital from investors for future development projects, which boost return on equity. As such, Westfield Holdings trades at price earnings ratio of 35-40 times, and manages A$22 billion of assets on behalf of the trusts.
Ideally, S-REITs should have proven a mechanism for restoring liquidity to the Singapore property market, by allowing assets to be sold at a valuation that would be higher than that achieved by a direct sale, although at a slight discount to the overvalued NAVs at which the buildings are booked.
As the biggest property company (formed from the merger of DBS Land and Pidemco) CapitaLand's failure is thus a major setback for the whole process.
It had looked the most likely to succeed, given the fact that rental yields on property malls at 5.75% were thought to be enticing enough to bring in the critical retail demand.
Office yields are much lower at 3-4% and from the outset were thus viewed as a harder sell. The only way to make office-based REITs more palatable would be for the property company to take a severe haircut on the net asset value at which it injected the building.
This was the dilemma that would have faced Keppel Land and its adviser, Merrill Lynch, which wanted to launch the first office-based S-REIT, injecting Capital Square, Prudential Tower and Ocean Towers.
That deal was already reckoned to be a tough proposition. With the failure of SingMall it looks doubly so.
The challenge will be to get the S-REIT market back on track as soon as possible. And if CapitaLand returns to the market, it may have to learn the lesson of this failure and structure the asset injection at a mild discount to the NAV it values the assets at.
Priced properly, S-REITs ought to fly. Moreover, from Singapore Inc's perspective, they need to.