Would Telkom be foolish to bet the farm on Cell C?

On paper, at least, it makes sense. Telkom would use its far superior balance sheet to buy a weaker rival, giving its mobile business instant scale. It tried to buy its (then larger) rival in 2017.

Telecoms, as we know, is a scale game. With 27 million subscribers, a combined operator will be better able to compete with MTN (28.9 million) and Vodacom (43.9 million). There are countless examples of markets around the world where four or five operators have consolidated down to three.

But the problem is that Cell C is arguably in worse shape today than it was two years ago when Telkom tried and failed.

Revenue and subscriber growth has flatlined – and earnings before interest, taxes, depreciation, and amortisation (Ebitda) for the 12 months to end-May 2019 (aligned with the year-end of Blue Label Telecoms) was down 19%.

In this period, Cell C’s net loss before tax was R3.8 billion. The fundamental problem, even post recapitalisation, remains Cell C’s debt. At that point, it was R8.9 billion, 19% higher than a year prior. Net finance costs were R2.15 billion in the 12-month period.

It is not a stretch to suggest that Telkom’s mobile service revenue for the full year to end-March 2020 will be higher than Cell C’s for the same period. Further direct comparisons are difficult because of the way Telkom has restructured its business (its tower portfolio sits under Gyro, and its core network, including fibre backhaul from towers, sits inside Openserve).

	Telkom (6 months to end Sep)		Cell C (12 months to end May)
		Change		Change
Mobile service revenue	R5.6bn	57%	R14.1bn	4%
Mobile subscribers	11.5m	76%	15.9m*	-2%
Prepaid subscribers	9.4m	93%	12.5m	-4%
Postpaid (contract) subscribers	2.1m	26%	1.1m	-6%
MVNO subscribers***	N/A	–	1.9m	13%
Prepaid ARPU****	R61	-13%	R48**	not disclosed
Postpaid ARPU	R178	-7%	R267**	not disclosed
Blended ARPU	R82	-21%	R74**	not disclosed

*Includes 413,000 ‘broadband’ customers

** Estimates, based on service revenue and active subscribers

*** MVNO: mobile virtual network operator

**** ARPU: average revenue per user

Any deal for Cell C has to resolve its ±R9 billion debt problem, which is ±R1.5 billion larger than it was last May. It didn’t add to its long-term borrowings in the year, but unfavourable currency movements ($) meant a R600 million swing. On the short-term front, it has relied on a bridge loan from vendor ZTE, capitalised finance costs, and a facility increase to stay afloat.

Read: Without mobile, here’s how much trouble Telkom would be in

Blue Label Telecoms paid R5.5 billion for 45%, and it doesn’t seem likely that a deal for Cell C will value the operator at anything more than R10 billion given the deterioration of its balance sheet and income statement since the recapitalisation.

Telkom group CEO Sipho Maseko may have some sort of ace up his sleeve.

Cell C has accumulated tax losses in excess of R20 billion. Could this be central to a deal?

Read: ‘It’s not MTN’s responsibility to save Cell C’

It is obvious that costs can be removed from Cell C, beyond those already taken out of the business since May. In fact, one could envision a scenario where, very simplistically, customers are migrated to Telkom’s network and practically everything else is shut down.

Beyond a subscriber base, it has brand equity, a store network (240-odd, many of which are franchises), a network in which it has woefully underinvested, and licensed spectrum. The two most valuable items to Telkom in this equation are the base and the spectrum.

Now that Telkom’s Mobile business is profitable, it needs to continue investing in capital expenditure to accommodate growth in subscribers.

Other expense increases in the business will be in line with revenues.

One way of looking at this is to calculate the amount of investment that is required in Telkom’s network to sustain growth in subscribers, and then to add to this the cost of acquiring customers via sales, marketing or incentives (and other overheads). Telkom knows exactly what this figure is in both the prepaid and postpaid space. It also knows the lifetime value of that active customer, i.e. when they become profitable.

At a hypothetical R10 billion, the cost per customer acquired from Cell C would be around R650. But the debt needs to be factored in. Also, Cell C says its total assets are valued at R18 billion (with R12 billion of this ascribed to its network).

On paper, there are many ways to make the deal seem appealing (one can be certain a number of investment bankers are concocting complicated structures to make a deal work).

Assuming a hypothetical price of R10 billion plus the debt, one might argue that, given the assets, the subscriber base could be ‘bought’ for R1 billion.

But Telkom would still have to assume Cell C’s debt, as a debt-for-equity swap seems far-fetched. Telkom could look to issue shares to raise capital, but neither this nor a debt for equity swap is likely to find favour with Telkom’s largest shareholders – government, directly via the Department of Communications (40.5%), and the Government Employees Pension Fund via the PIC (11.9%).

The assumption of Cell C’s debt, despite Telkom’s protestations that it has a “healthy” balance sheet, certainly stretches things to ‘interesting’ levels. And this doesn’t factor in Telkom’s guidance which already sees net debt/Ebitda reaching 1.5 times in the next two years.

Add in a hypothetical R10 billion price tag for the acquisition (also funded by debt), and things start looking rather uncomfortable.

	Telkom	Cell C	Combined entity	With acquisition debt**
Net debt*	R11.775bn	R8.235bn	R20.01bn	R30bn**
Ebitda	R10.08bn (for FY)	R3.391bn	R13.471bn	R15bn**
Net debt/Ebitda*	1.2x	2.4x	1.5x	2x

* Excludes impact of IFRS 16

** Estimates

Telkom has some very serious issues to resolve in its own business, as detailed on Tuesday (read: Telkom is out of time). Its cash burn, negative free cash flow, and the reliance on debt to fund capex are all big warning signs.

Perhaps whatever Sipho Maseko has seen when looking over the cliff edge has him horrified. Executives have surely modelled out the decline of the fixed line business. Numbers from these six months are really just the start. It is absolutely cratering and there’s no turning back.

	March 2018	Sept 2018	March 2019	Sept 2019
Mobile subscribers	5.2m	6.5m	9.7m	11.5m
Growth	21%	25%	49%	19%

Until now, mobile has made up for the declines elsewhere but, worryingly, growth in mobile subscribers has slowed sharply in the last six months. BCX is going nowhere and fibre is not making up the losses elsewhere in its fixed line unit, nor is it large enough yet to really move the needle. This has possibly forced Telkom’s hand. 

The question Telkom – and Sipho Maseko – must answer is where growth will come from.

That might explain why it’s back at Cell C’s door …

* Hilton Tarrant works at YFM. He can still be contacted at [email protected].