When reports surfaced last year that China Mobile was interested in buying Digicel’s Papua New Guinean and South Pacific business, it didn’t go down well in Canberra. 

China Mobile is, like all big Chinese companies, an agent of the Chinese state. By buying Digicel’s South Pacific business, the fear is it would be able to listen in on conversations with local leaders, neighbouring countries, and visiting Australian officials. The Australian State also worried the business would be used to bribe local officials.

China aspires to be a superpower, and any would-be superpower needs a sphere of influence. Like the USSR in Eastern Europe or the USA in Latin America, China wants to call the shots in East Asia and the South Pacific. 

China is throwing its weight around more assertively. It has laid claim to the entire South China Sea and is building artificial islands there for military bases. It’s also bribing and buying influence, by for example investing in neighbouring countries’ infrastructure. 

Papua New Guinea was directly ruled from Canberra from 1915-1975. It’s about 150 kilometres from the coast of Australia, not much farther than Ireland is from Holyhead. The Australians see it as their backyard.

So the Australians have devised a plan. Digicel’s Papua New Guinean business is to be bought, not by a Chinese company, but an Australian one. Telstra, a big Australian telco, has lined up a bid for the assets. And an investment arm of the Australian government, Export Finance Australia, has quickly had its powers amended to permit equity investments in “certain circumstances”. The Australian government would reportedly finance $1.5 billion of the $2 billion deal.

It’s a happy circumstance for Denis O’Brien. His South Pacific business is the centre of attention. And it couldn’t have come at a better time. If the Australians can be convinced to pay a good price for the business, it opens a path for O’Brien to finally get Digicel’s creditors off his back. 

The machete-through-the-jungle years

Not many people look at Papua New Guinea and see opportunity. 

The country is split into more than 600 islands, choked with jungle and vegetation. The islands are tropical and they’re mountainous – the range that splits the island of New Guinea rises to over 15,000 feet, about as tall as the Alps.

The people there have been dealt a tough hand. There are no native crops suitable for cereal farming, no native animals suitable for farming, and no large game to hunt. By the shoreline there are fish. But in the interior of the country, people have relied on farming sweet potatoes, imported from South America by the Spanish. Sweet potatoes help with calories, but not protein. A shortage of sources of protein has bedevilled the people of New Guinea for forty thousand years, says anthropologist Jared Diamond. 

With no native cereals and no native livestock, farming in Papua New Guinea is hard work. It’s all hands to the pump. 

Because of this, Papua New Guinea is the most rural country in the world, with only 13 per cent of the population living in towns and cities. In rural parts of Papua New Guinea, “most people live almost as they have for tens of thousands of years”, said Dr Amanda Watson, an expert on Papua New Guinea at the Australian National University. It’s also one of the poorest countries in the world, with a GDP of $2400 per capita.

Given all these problems, what attracted Denis O’Brien and Digicel to Papua New Guinea? Well, the country may be poor, but it isn’t small. It has a population of 8.7 million. And crucially, it was virgin territory for a cellular network. When Digicel arrived in 2006, there was no incumbent.

Cellular networks are sometimes called natural monopolies. That’s because building the network requires a big upfront investment. The returns to the first company to build a network are big because they get the market to themselves. But returns to each subsequent network are smaller because it has the same upfront costs but a smaller payoff. The natural monopoly effect is particularly pronounced in Papua New Guinea, where bad roads and tough terrain make it very expensive to build cellular towers. 

“There was a huge amount of cost for them when they initially set up this network,” said Dr Watson. “In many places, they had to construct a road to be able to put in a tower. Or they had to helicopter in the parts. Or pay local people to hand-carry the parts across footbridges.” 

This is Digicel’s bread and butter. It goes into developing nations where the local competition is either of a poor standard or non-existent, and it makes huge upfront investments to build out cellular networks. Once the network is in place, Digicel dominates the market and can charge handsomely for its services. Digicel has 91 per cent market share in Papua New Guinea. It has the number one market position in 25 of its 32 markets. 

It might be that Digicel’s services cost more than they would in a rich competitive market like in Europe. But that’s not to say the customers were getting gouged. Digicel made investments that wouldn’t have otherwise been made. 

“People were delighted to be able to be able to hear the voice of an immediate family member who was working in another part of the country, whose voice they hadn’t heard for years,” said Dr Watson.

“I remember in about 2005 or 2006, before Digicel, seeing people queuing at public phones. People would sometimes spend all day, or longer, travelling from the village to go to town to make a phone call. So the significance of Digicel when it came in — it was almost like a saviour.”

Digicel ran this playbook over and over the Caribbean and South Pacific. Not every country was virgin territory for a cellular network, like Papua New Guinea was. But in most of its markets the incumbent companies offered a shabby service, which created an opening for Digicel to come in and take over. 

The following chart shows where Digicel makes its revenue, and the overall trend from 2018 to 2020. The biggest five markets are in Jamaica, Haiti, Trinidad & Tobago, El Salvador and Papua New Guinea. Those markets accounted for 62 per cent of Digicel’s 2020 revenue. The remaining 38 per cent is made up of a long tail of 27 smaller markets: Anguilla, Bermuda, Curacao, French Guiana, Nauru, St Kitts and Nevis, Suriname, Vanuatu, Antigua and Barbuda, Bonaire, Dominica, Grenada, Panama, St Martin and St Barth, Tonga, Aruba, British Virgin Islands, Guadeloupe, Martinique, St Lucia, Barbados, Cayman, Fiji, Guyana, Montserrat, Samoa, St Vincent and the Grenadines, and Turks and Caicos.

Building a network from scratch in 32 tiny markets scattered around the Caribbean and South Pacific is hard work, as we've seen. But the reward is a richly profitable franchise, with little competition. Digicel and its main Caribbean competitor, Liberty Global, each have 45 per cent Ebitda margins. That's compared to an Ebitda margin of 27 per cent for Europe's Vodafone or 36 per cent for America's AT&T. 

Another advantage of the Digicel business model is that cash flows are pretty stable. Digicel's earnings don't vary a lot because Digicel doesn't have to worry about price wars with rivals, or a sudden collapse in demand for its services. From Denis O'Brien's perspective, the 92 per cent owner of Digicel, this is great because it has allowed him to fund the company with debt. 

By loading Digicel up with Debt, O'Brien has been able to extract $1.9 billion from it in dividends. The dividends were O'Brien's reward. But the debt ended up nearly costing him the business.

Skin of the teeth

To deal with Digicel's heavy debt burden, in 2015 the company came up with a sensible plan: it would list on the stock market and use the proceeds to pay down the debt. O'Brien's ownership of the company would be diluted somewhat, but Digicel's debts would be brought under control.

Digicel spent six months working on the deal. The plan was to raise $1.7 billion for 39 per cent of the company. Of the proceeds, $1.3 billion was to be used to cut the debt.

At the last moment, O'Brien pulled the float. He wasn't happy with the offer price. He decided Digicel would stagger on with its heavy debt burden and come back to the public markets in future, when conditions were better.

In the following years, conditions never got better for Digicel. Its debt burden got even bigger when Digicel's business hit a rough patch. Digicel was hit by the decline in phone calls, which are profitable, at the expense of voice over IP, which is less profitable. In addition, Digicel had borrowed in dollars, yet it earned money in local currencies. In a 2019 note the credit rating agency Moody's said:

"Local-currency depreciation against the US dollar has reduced Digicel's revenue by several percentage points annually in recent years, with Digicel having only about 40% of its revenue in US dollar or in currencies pegged to the dollar. In addition, there is a mismatch between the company's largely local-currency-denominated revenue and dollar-denominated debt."

For example, Haiti is one of Digicel's biggest markets. In 2015, 100 Haitian gourdes bought $2.10; today it buys $1.04. 

Between the decline of its core voice telephony business and a weakening of key local currencies, Digicel found itself in trouble. Its debt ratios got worse and worse, peaking at around 7.5 times Ebitda in early 2020. 

O'Brien pushed back at Digicel's creditors, renegotiating some debts and extending their maturities in 2018. This bought Digicel time. But a day of reckoning was due on 15 April 2021, which is when a giant $1.3 billion debt payment came due. Unless Digicel could find a way to refinance the $1.3 billion, there was a real risk of O'Brien losing the company. In December 2019 Sul Ahmad of Fitch, a rating agency, said:

"What we're saying is that there's definitely a strong likelihood of default at those levels. Because operating performance could improve between now and then – but not likely enough for the company to be able to refinance those debts at those levels."

The turning point came in April 2020. That's when Digicel and the creditors came to a deal to write down $1.7 billion of Digicel's $7 billion debt pile, 24 per cent of the total. Digicel was able to convince the creditors into accepting the write down because the alternative was chasing the company down through the Caribbean courts system, which has never before handled a corporate bankruptcy of Digicel's size. The April 2020 deal opened a path for O'Brien to keep full control of the company — subject to a few conditions.

Remember that the deal was agreed in April 2020, which was about the single scariest month in the last two years. A big condition of the deal was that Digicel committed to refinancing a $920 million bond that came due in March of 2023. If Digicel wasn't able to refinance that bond, Digicel's creditors would get 49 per cent of the equity. At the time, it wasn't a given that this would be possible, because even after the writedown, Digicel was considered highly levered.

In the last 18 months, things have gone Digicel's way. First, the world's central banks have kept the flow of money circulating in the world economy. That's been good for bonds in general. Good news for bonds in general is good news for Digicel bonds in particular, and makes them easier to refinance. Second, Digicel's underlying business has performed better. Third, foreign exchange has moved back in its direction. 

Now, the goings-on between China, Australia, Papua New Guinea and the South Pacific are working out to Digicel's advantage.

The new deal

Telstra is reported to have offered just short of $2 billion Australian dollars for the South Pacific business, or $1.49 billion. What does this do for Digicel?

Right now, Digicel has $5.55 billion of net debt, and about $930 million of Ebitda. Divide one by the other and you get a net debt to Ebitda ratio of 5.9. 

The South Pacific business generates 18 per cent of Digicel's Ebitda, meaning after a sale it would have $4.06 billion of net debt and $762 million of Ebitda, for a new debt to Ebitda ratio of 5.3. It would be "modestly deleveraging", says Sul Ahmad of Fitch. But its impact on Digicel's creditworthiness would be bigger than just losing half a turn of leverage. It would show that Digicel is a grown-up company, serious about deleveraging, with more in its toolkit than hardcore restructuring negotiations. 

Ahmad explained why Fitch would see this deal as positive from a ratings perspective: “We take a holistic view of the credit story. It’s not solely that this transaction could moderately improve leverage and coverage ratios. In addition, it shows a willingness to slim down the business and expedite asset sales, which would improve our assessment of corporate governance.”

O'Brien cares about hanging onto his equity, and the biggest threat to his equity is the convertible note that kicks in in 2023, in the event of Digicel failing to refinance the $920 billion bond. This deal helps because it slightly improves Digicel's debt ratios, and as we've seen, makes Digicel a generally more creditworthy business.

But it's possible, if this transaction raises enough money, for O'Brien to buy out the convertible loan note. Covenants attached to the note say a $940 million secured bond and $400 million unsecured bond must be cleared before the convertible loan note, which is worth $200 million. That sums to $1.54 billion. So with $1.49 billion from a South Pacific transaction, O'Brien wouldn't be far from clearing that most troublesome part of Digicel's capital stack. 

Say the deal goes through. Where would it leave Digicel? 

The company's hacking-through-the-jungle days are behind it. And if the Telstra deal goes through, so might be its teetering-on-the-edge-of-bankruptcy days. With the debt back under control, Digicel could settle in as the dominant telco in 25 small markets, and a cash cow for its owner.