A Hong Kong listing should solve a currency problem first

During 2025 a large number of companies already listed in Shanghai and Shenzhen announced plans to list in Hong Kong as well. Some had clear reasons: overseas plants to fund, foreign customers to reassure, acquisitions abroad that needed foreign currency. Others seemed to be following a crowd. In August our board decided to defer a Hong Kong listing that our chairman had proposed in March. I recommended that decision, and I want to set out the reasoning, because I think it applies to many mainland consumer companies.

My position is simple to state. For a consumer company that earns and spends almost all its money in renminbi, a second listing in Hong Kong is justified only when the company needs foreign currency for a use that earns more than it costs, or needs shareholders it cannot reach any other way. Visibility, prestige and the fact that peers are doing it are not reasons. They are what a listing feels like from the inside.

What a second listing actually buys

A Hong Kong listing does four useful things. It raises money in Hong Kong dollars, which can be used offshore without the approvals that moving renminbi abroad usually involves. It brings in international institutions, who tend to ask harder questions about governance and capital allocation. It gives a company a share that can be used to pay employees or vendors outside the mainland. And it raises the company's profile with partners in Southeast Asia and beyond.

Cuiling's chairman made the strongest version of this case. He proposed raising about HK$2.5 billion, mostly to buy 60 per cent of a New Zealand processor of milk powder and butter, and the rest to fund imports. The acquisition would have secured a source of imported dairy at a time when he expected domestic milk to tighten again.

The question I asked was what we would do with the currency. Our dollar purchases had fallen from USD 71 million in 2022 to USD 20 million in 2025, because we were using our own milk first. Ninety-eight per cent of our revenue is in renminbi. The New Zealand processor, on our analysis, would have earned less than our cost of capital at 2025 powder prices, and it would have added a New Zealand-dollar exposure to a balance sheet that had almost none. The listing would have funded an acquisition whose main justification was that the listing made it possible.

International investors are a better argument, and I underweighted it at first. They do bring a different discipline. But they can already own our A-shares through Stock Connect, and several do. The discipline they bring depends more on how well a company discloses and engages than on where its shares trade.

The discount and the calendar

The costs are easier to count. The first is valuation. For many years the H-shares of mainland companies have tended to trade below the A-shares of the same company, sometimes by a wide margin. Our advisers' range implied an H-share price 25 to 35 per cent below our A-share price. Raising HK$2.5 billion at that price would have meant issuing roughly 9 per cent more shares, selling part of the company at a discount to its own market value in Shanghai.

There is also the question of who bears that discount. The controlling shareholder of a state-linked company may value a listing for reasons that minority shareholders in Shanghai do not share, and it is those minority holders whose stake is diluted at the lower price. A board should hear that argument stated explicitly, and its independent directors are the right people to ask for it.

The second cost is the calendar. A dual-listed company answers to two exchanges, publishes price-sensitive information simultaneously in both markets and maintains investor relations in two languages and two time zones. An offering abroad by a mainland company also requires a filing with the CSRC, which adds time and uncertainty before the market window can be used. Our estimate of the ongoing cost was about RMB 22 million a year in fees, disclosure, translation and investor relations, before management time.

The third cost is harder to price: the effect on management's attention. I prepared a Hong Kong listing for a beverage company in 2015 that was withdrawn when markets fell. For most of that year the finance team, the chief executive and the founder worked on the offering rather than the business. Much of the cost was spent before anyone knew whether the listing would happen.

When I would say yes

None of this makes me opposed to Hong Kong listings in principle. Our board asked me to bring an annual capital-structure review, and we agreed three conditions under which the question would return: an offshore acquisition above RMB 1 billion, dollar purchases above USD 60 million a year, or a valuation gap between our A-shares and an H-share price narrower than 15 per cent. Any one of those would change the arithmetic.

For other companies, I would say yes to a second listing when the company has a funded plan for offshore money that earns more than its cost; when its foreign-currency costs or revenues are large enough to need a natural hedge; or when it has international employees or partners who should own its shares. I would say not yet when the main argument is the window, the peer group or the prestige. Timing matters less than it seems. Windows open and close, but a company with a real use for the money will usually find a price at which the issue makes sense; a company without one will find that no price does.

We did not simply say no. Instead of buying the processor, we signed a five-year supply agreement with it in November 2025, with the option of a minority stake later. And because the chairman was right that international investors would have improved our discipline, we began publishing fuller English disclosure and meeting the foreign holders of our A-shares directly.

For directors weighing the same proposal, three questions will do most of the work. What will the foreign currency be used for, and what will it earn? What price, relative to the A-shares, would make the issue worth doing? And how much of management's next eighteen months will the listing consume? If the answers are vague, the listing is answering a question nobody has asked.