In my last note on Naspers dated 6 December 2016 (see “Discount to value of Tencent stake widens”) I discussed the phenomenon of the market capitalisation of Naspers steadily descending from a substantial premium to the see-through-value of the Tencent stake to a substantial discount. However, I cautioned that this was not necessarily an opportunity to pick up a heavily discounted entry to Tencent. This latest note further unpacks this theme of value detraction and graphically depicts the deterioration in four years. On a sum-of-the-parts basis, the intrinsically valuable Pay TV business, of which MultiChoice is the flagship, is not being reflected whilst ecommerce is value detracting as returns remain elusive after years of heavy spend. The $3,3 billion due from the sale of Allegro is not being credited in the valuation. Arguably, shareholders would be better off with an unbundled and separately listed MultiChoice, a Naspers as an investment holding company with its only asset being Tencent, and a newco with development assets in the portfolio hived off too. For investors in Naspers, the material discount to Tencent provides a cushion for possible negative surprises and some headroom for future upside if ecommerce is a positive surprise. But as time has gone on, the growing discount is a telling metric on receding confidence by the market in Naspers’s ability to plough its own furrow. Naspers today is essentially a marked-down Tencent proxy. Take a listen here.
At the beginning of January 2012, five years ago, Naspers cost you R360 per share. At the prevailing exchange rate, the equivalent in dollars was $47 per share.
At that time, Tencent, in which Naspers has a 33,51% interest, cost HK$31 per share. The Hong Kong dollar shadows the US dollar at a rate of around HK$7,75 and so Tencent cost $4 per share.
As we begin a new year, a Naspers share will cost you R2150 (an Easy Equities fractional ownership option makes this eminently affordable). That translates to around $157 per share.
So, in rand, your Naspers share over five years is up approximately six-fold or 500% whilst in dollars it is up more than three-fold or about 230%.
Not bad you may think, not least given that the JSE All Share Index is up approximately 50% in rand over five years and broadly flat over almost three years and over one year. So, a passive index tracking tactic wouldn’t have done particularly well versus inflation in recent times. In dollars, the JSE is down about 12% over five years, excluding dividends.
But let’s dig a bit deeper because conventional thinking has it that Naspers is the tail wagged by the Tencent dog. If that were true, Naspers would simply reflect the Tencent performance. However, nothing could be further from the truth.
That Tencent share today is HK$198 which is over $25. In US dollar, Tencent is up over six-fold this past five years or 530%.
The following graphs illustrate the divergence between Naspers and Tencent referred to in the previous note.
Naspers and Tencent based to 100 over five years in US dollar at prevailing exchange rates for the ZAR and HK$
In the first graph, each share price data point is converted to USD at prevailing rates of exchange on ZAR and HK$. In turn, I have formulaically based both Naspers and Tencent to 100 commencing January 2012. Standardising on the US dollar is appropriate because Naspers reports financial results now in US dollars, most of Naspers earnings is international, notably China, and the Hong Kong dollar shadows the USD and has been broadly unchanged over the period.
Ratio of the Naspers share price in USD to the Tencent share price in USD
In the second graph, a ratio is calculated between the Naspers share price in USD and the Tencent share price in USD. The ratio between the two has narrowed by half – in January 2012 Tencent was 8% of the Naspers USD price whereas it is now 16% or to put it another way going from a ratio of 12x to 6x.
Naspers market capitalisation in US dollar as a percentage of the value of its investment in Tencent in USD
At the end of December 2012, the market capitalisation of Naspers was R225 billion which translated to $26,5 billion. Tencent had a market capitalisation of HK$463 billion which translated to $60 billion. The Naspers share of that $60 billion was thus $20 billion.
Therefore, as at December 2012 the market capitalisation of Naspers was 133% of its share of Tencent or put another way Naspers traded at more than 30% premium to the see-through-value of the Tencent stake, normalised to USD for both stocks. Graph #3 above illustrates this.
What this meant was that around $6,5 billion or R55 billion in value at that time was attributable to operations other than Tencent. This suggested that the market priced in value to Naspers over and above its investment in Tencent.
This premium subsequently degraded to broad parity, as is seen in graph #3. This means the value of businesses other than Tencent were increasingly ascribed diminishing value.
Throughout 2016, the market capitalisation of Naspers traded below the see-though-value of its Tencent stake of 33,51%.
At the end of December 2016, the Naspers market capitalisation was R884,1 billion which translated to $64,3 billion.
The market capitalisation of Tencent at the end of December 2016 was HK$1,8 trillion or $232 billion of which Naspers shared in $77,7 billion.
The difference between the Naspers market capitalisation of $64,3 billion and the value of its Tencent investment of $77,7 billion is a negative $13,4 billion or a 17% variance. Put another way, the value of the Naspers share of Tencent was 20% more than the Naspers market value.
Depending on currency and share price movements, Naspers now trades 15% to 20% less than the value of its Tencent stake.
In broad terms, there has been a 50 percentage points movement over five years – a 30% premium to a 20% discount. This explains why Tencent has strengthened from 8% of the Naspers USD share price to 16% or Naspers weakening from 12x the Tencent share price to 6x. In dollars, this negative variance is $20 billion.
What this implies is that Naspers is taking away value rather than adding value.
On a sum-of-the-parts basis, the intrinsically valuable Pay TV business, of which MultiChoice is the flagship, is not being reflected whilst ecommerce is value detracting as returns remain elusive after years of heavy spend.
Even if the market were pricing in a value on MultiChoice, it means that the ecommerce operations are even more value negative than the roughly $13 billion differential between the Naspers market capitalisation and the value of the Tencent stake.
It is inconceivable that Tencent is being discounted through Naspers to this extent. Naspers is not a pure investment holding company like Reinet or a Remgro, both of which typically trade at varying discounts to declared underlying net asset value. Moreover, the fact that value used to be ascribed to assets other than Tencent is testament to the fact that investors expected Naspers to unearth another Tencent or develop MultiChoice or find the next big thing in online retail.
The Tencent share price was under pressure this past quarter, as depicted in graph #4 – not because of the company per se, whose fundamentals are unaltered, but because of pressure on China equity and an exiting from China of exchange traded funds. This is entirely China-macro related.
Tencent share price in HK$
Whilst the stock has recovered to HK$198 from a low of just below HK$180 before Christmas, it is still down from an all-time high of HK$218 at the end of September.
Even at this lower level the market cap works out to US$240 billion. Tencent is the largest constituent of the Hang Seng Composite Index – in the region of 9,6%. The next largest is HSBC at 9% and then AIA the insurer at 5%.
Despite Tencent being such a heavyweight, Information Technology as a sector is 12% of the Index compared with Financials at 38%, the largest.
Fund redemptions automatically weigh on a stock like Tencent because of its sheer size and dominance in the IT sector.
Judging from the firmer share price of late, fundamentals are seemingly reasserting with respect to Tencent, notwithstanding China jitters.
The Chines yuan too plays a part and affected by US factors such as interest rates, bond yields, and Trump trade politics. Back in April 2016, the yuan was below CN¥6,50/$ and today is close to CN¥7,00/$.
There could conceivably be further China equity/yuan pressure after the US inauguration on 20 January. If so, Tencent won’t be immune but ultimately its fundamental advantages and good profitability should prevail.
If we assume that the China sell-off has travelled much of its course this is positive for Tencent and by implication Naspers. However, Naspers still has to be looked at on the basis of what it can add to value creation, independent of Tencent.
The self-professed aim of Naspers is that “we build leading companies, we’re entrepreneurs, we push for performance in everything we do, we back local teams and learn from each other, we’re nimble and seize opportunities”.
Naspers also reckons “we win” by focusing the portfolio around core leadership positions, developing and growing core platforms by leveraging leadership positions, rolling out the most successful platforms to new, adjacent business models, expanding into new geographies and transforming by making investments into new, disruptive platforms.
What is clear to the external observer is that developmental assets, largely ecommerce, are deteriorating in their Group contribution relative to Tencent and are cash absorbing. Furthermore, it is difficult to see these assets reaching sizeable international scale.
If leverage is the in thing, why get rid of Allegro, one of the better assets? Allegro, the Polish online marketplace, is being divested for a total of $3,253 billion, which should reflect on the March year-end balance sheet. Naspers has not followed on recent funding raises for Flipkart or Souq, diluting both interests.
The only relatively positive thing of late was ibibo being merged with MakeMyTrip in exchange for issuance of new shares by MakeMyTrip and thus giving greater scale in travel booking in India. Naspers ends up with 40% of the merged entity. Detail on the deal is sketchy but I estimate the value attributed to ibibo at $700 million or R23 per Naspers share.
The key to the value detraction but also possible value unlock is so-called “development spend”, a term often used in ecommerce and which is code for losses in conventional accounting. Development spend is not an IFRS measure.
Naspers companies spent $387 million alone in the first six months of the 2017 financial year and will spend about $600 million for the full year. In the six years to March 2016 about $3 billion was spent and in the five years 2017 to 2021 some $1,5 billion to $2 billion is possible. That could result in up to $5 billion in eleven years or an average of $450 million per annum. This is shown below in graph #5.
Tencent, by way of comparison, is very profitable and will make an estimated EBITDA of $9,4 billion and after tax earnings of $6,5 billion for the year to December 2016. Capex though is $1 billion.
Development spend in USD million – 2017 thru 2021 estimates
How does this development spend get paid for? Mostly by MultiChoice profits and a Tencent dividend.
I estimate that the Tencent dividend for the year ending December 2016 will be approximately $700 million of which Naspers gets $235 million. MultiChoice will contribute over $400 million in after-tax earnings in 2016, despite a soft rand. Together, that is over $600 million from just those two sources.
Profitable ecommerce entities within Naspers made $288 million for the year to March 2016 and $215 million for the six months to September 2016. Allegro, OLX and Avito probably constitute the bulk of the profits with Allegro, now sold, the largest. The cash value of Allegro is over R100 per Naspers share.
With over $200 million of profit from Allegro walking out of the door to be replaced with $3,3 billion in cash from the sale, not only is ecommerce operating profitability materially diminished but there are question as to how well the cash received will be spent. Is it more expensive mud thrown at the wall with the hope some sticks? Whilst development spend in the older book is decreasing it is countered by increased spend elsewhere.
To get a decent internal rate of return on the billions of dollars spent will require Naspers to generate at least a $1 billion in EBITDA in ecommerce from 2021 onwards.
It is not surprising that the large discount to the Tencent valuation has coincided with growing development spend, little or no financial disclosure and so far, elusive returns.
My latest sum-of-the-parts adjusts for the sale of Allegro and includes the cash received. The remaining portfolio is therefore reduced to $5,6 billion on a present value basis. That means an effective value for ecommerce of approximately $9 billion, including Allegro.
However, MultiChoice I value at around $9 billion on a 100% basis.
I have a sum-of-the-parts value of R3018 per share at the current rates of exchange. This includes Tencent at R2566 and all the other assets at a value of R515 per share, net of debt. If I apply my usual 20% fair value discount, to provide a cushion for error, the SOTP drops to R2465 per share.
My SOTP value on Tencent remains at HK$198 per share.
Arguably, shareholders would be better off with an unbundled and separately listed MultiChoice, a Naspers as an investment holding company with its only asset being Tencent, and a newco with development assets, notably ecommerce, hived off too.
Such a demerger and a clearer line of sight on the financials would conceivably unlock value. MultiChoice alone could potentially be R100 billion or so JSE listing.
The chances of such a corporate action happening are almost certainly zero – at least for now unless shareholder pressure mounts.
Whilst Naspers shareholders have done well from a capital appreciation perspective, shareholders have not done as well as they could have given the increasing value diminution relative to Tencent.
In some respects, there is a bit of money illusion – you’ve done ok measured in rand, but that is only one side of the ledger.
For investors in Naspers, the material discount to Tencent provides a cushion for possible negative surprises and some headroom for future upside if ecommerce is a positive surprise. But as time has gone on, the growing discount is a telling metric on receding confidence by the market in Naspers’s ability to plough its own furrow. Naspers today is essentially a marked-down Tencent proxy.
Trading Buy and Portfolio Buy maintained.
Naspers in ZA cents
Wishing you profitable investing, until next time
Mark N Ingham