Temasek Holdings’ Returns are Not Special

Temasek Holdings began publishing its annual reports, only from 2005.  It is a welcome move towards transparency, although much more needs to be done.  Working with what we have, the financial reports make interesting reading, but throw up many questions.

For example, in the 2010 annual report, Temasek Holdings’ claimed 20-year ROI was 16%.  In the 2014 annual report, it was 6%.  This is a substantial drop.  Using data available in the public domain, when we calculate backwards, we arrive at the conclusion that the early 1990s was the period of greatest growth for Temasek Holdings.  This is important.  In 1993, SingTel was listed.  But in 1992, and 1994, Temasek Holdings did not list its portfolio value, which is not necessarily a coincidence.  Again, calculating from data, the listed portfolio value was approximately S$15 to 20 billion.

The end of Temasek Holdings’ fiscal year was moved to the 31^st of March, from the 31^st December, in 1993 to align it with the Singapore Government and GLCs.  On the 31^st March 1994, portfolio value was estimated at S$70 billion.  This is a marked increase in a period where there was no evidence of major portfolio activity, or market movement, to account for this.  We should note that Temasek Holdings has never actually provided any meaningful breakdown of its portfolio, let alone the value of its position in SingTel.

Going back to 1993, the year before SingTel went public, Temasek Holdings’ listed holdings were valued at S$6.5 billion.  This means unlisted portfolio assets were valued between S$8.5 to S$13.5 billion.  It would be more likely that the actual number is closer to the former than the latter.  Since Temasek Holdings listed its portfolio value at S$70 billion in 1994, and considering the S$8.5 billion, this would mean SingTel was valued at S$17.5 billion after listing.  This is a more than a 100% increase, which is reasonable.  SingTel was a grossly undervalued asset, considering its dominant market share, its captive client base and assets.

The conundrum here is that if we calculate the value of other known holdings, we arrive at a portfolio value that is S$27.5 billion.  This is still less than the claimed portfolio value of S$70 billion, in 1994.  I would assume that Temasek Holdings valued SingTel at a nominal sum, which would make sense, since when the Singapore Government transferred SingTel over to Temasek Holdings, it was without compensation from Temasek Holdings.  It could be argued that they gave a public asset away, as a form of capital injection.  Considering the 1994 portfolio is estimated at S$17.5 million or so, it implies that the remaining S$10 billion was now S$42.5 billion.  This would further mean an asset growth of over 400%, which is unexplainable.

Based on the scarce information we have then, we can calculate that the declared portfolio of Temasek Holdings have earned a 12.5% return since 2005, the year they began publishing their reports.  However, their undeclared holdings have lost just over 9% per annum in that same period, until 2018.  If we consider that the averaged total return of their listing history, it was just over 10% per annum, but the earnings per share growth is 7% in that same period.  There is a discrepancy in that number since the total earnings have exceeded their market earnings.

Now, to compound the issue, we know, for a fact, that in 1997, SingTel was given S$1.5 billion for relinquishing its monopoly on telecommunications.  Since the owners of SingTel, M1, and Starhub are ultimately the same, it is arguable if there really was a relinquishment of monopoly.  This was just four years after its listing.  This S$1.5 billion was 20% of its market capitalisation then.  This money was a subsidy in all but name.

This method was later used on Temasek on SMRT, on Singapore Power, on ComfortDelgro, on Singapore Airlines.  Public assets were transferred to profit centres at nominal cost, or in an exercise that was essentially a capital injection.  I can understand the rationale behind this.  The idea was to make them efficient by subjecting them to market forces, and lower costs in the long term.  My point is that they were hived off to Temasek Holdings at no capital acquisition costs.  Temasek Holdings then brought them to market at a significant gain, and claimed these gains as part of portfolio growth, artificially inflating their returns.  Once we account for this, Temasek Holdings is not better than any other asset manager.  It is not special.