Haier Smart Home: The failed arbitrage play
I frontload this blog-post with the disclaimer that the following is nothing that even remotely resembles investment advice, neither for going long the stock, nor for shorting it; it’s a purely theoretical reflection without real-world application. And with that out of the way, let’s take a look.
The German and Chinese governments initiated the project of the China Europe International Exchange (CEINEX), a joint venture between the Shanghai Stock Exchange (SSE), the Deutsche Börse Group (DBAG) and the China Financial Futures Exchange (CFFEX). The CEINEX brought a new phenomenon into being: Chinese D-shares; with D like “Deutschland” or “Déguó” (‘Germany’ in Chinese). As a financing channel, the D-share market was supposed to reduce the cross-border capital flow risk and exchange rate risk for Chinese companies. So far the idea. But the reality is that CEINEX, founded in 2015, to this day has only seen one company list in D-Shares on its exchange, Haier Smart Home, one of China’s largest home appliances manufacturers (refrigerators, washing machines, air conditioners…).
The Haier D-share listing is subject to the German Prime Standard, and Haier has complied with this standard by delivering the financial reports and data required since its IPO in 2018. For investors, the dividend has been transferred accurately each year.
Completely apart from the D-share project, currently companies from China are speeding up their foreign listings by leveraging Global Depositary Receipts (GDR), with a preference for a listing in Switzerland.
The Haier D-share is not a GDR, or ADR (American Depositary Receipt), for that matter. A GDR is a certificate constructed and issued by banks; it is not an ownership stake in a company, but the receipt of an ownership stake. To the contrary, the Haier D-shares are proper fractional parts of the company Haier Smart Home.
With ADR/GDR there is the legal risk that this currently tolerated structure will someday not be tolerated any more by the Chinese state. In truth, while a risk, it seems very unlikely to materialise, as then China would lose access to the Western capital markets.
There is the possibility that both concepts, i.e., the D-shares and the GDRs, will continue to run side by side, but for that, there would have to be more D-share IPOs in the medium term. It is questionable if the D-share construct will be maintained because of a single share on the CEINEX, especially when many Chinese companies are now deciding for a GDR listing.
Taking a closer look at Haier Smart Home: they have three classes of shares. A-shares, traded in Shanghai; D-shares, traded in Germany; and H-shares, traded in Hong Kong. The D-shares were created in 2018, as mentioned above; the H-shares in 2021. Haier maintains that the three share classes grant investors the same economic and ownership rights. Haier Smart Home’s total number of shares outstanding, including all three classes, is roughly 9235 Mio, whereof ca. 67% are A-shares, ca. 30% are H-shares and ca. 3% are D-shares, the latter with ca. 271 Mio shares.
Currently, Haier Group, the parent company of Haier Smart Home, holds 21.45% of all D-shares via its wholly owned subsidiary Haier International Co (HIC), and the Silk Road Fund holds 19.93%. And yes, this Silk Road Fund is the same state-backed investment vehicle that is often mentioned in the media. This underscores that the IPO of a Chinese D-share in 2018 was supported by a strong political motivation, and it is political motivation that keeps the D-share-class alive. Ownership of D-shares by the Silk Road Fund has been relatively stable. There was a small reduction in the position at the end of January 2021 by just under 1 million shares in an off-exchange transaction at 2.333 EUR per share. The number of shares bought by HIC was exactly the number of shares sold by the Silk Road Fund, hence, it can be assumed that HIC bought directly from the Silk Road Fund.
The failed IPO
The issuance of the D-Shares was intended to finance Haier’s acquisition of Candy, an Italian domestic appliances manufacturer.
At the IPO, the D-shares shares were sold at a significant discount to the A-shares in Shanghai. In itself, selling shares below value seems illogical, but it is everyday life for Chinese companies that often issue H-shares (Hong Kong) at a discount to A-shares (Shanghai). They do so because they want to gain reputation with the international Hong Kong listing and be able to make an international takeover against shares if necessary.
Still, Haier at the D-shares IPO could only sell shares at bargain prices, and hence the IPO can be called a failed enterprise. Haier originally wanted to raise about 1 billion EUR from the placement. But in fact, they ended up issuing ca. 271 Mio shares at a price point of 1.05 EUR. Even for that price, buyers were scarce, hence, IPO-helpers jumped in, basically banks, suppliers or other institutional investors, who took good chunks at the IPO price. Accordingly, at one point, “Industrial and Commercial Bank of China - Financial Planning Agent” held 9.74% of the total number of D-shares, “China Investment Corporation” held 8.86%, “Shougang Holding Trade (Hong Kong) Limited” held 8.78% and “Rechi Precision” held 7.03%. And from the outset, the Silk Road Fund has held the approx. 20% it has basically maintained so far.
Among the names of the IPO-helpers given above, Rechi Precision is of particular mention. They are Haier’s compressor supplier and have their supplier factory in Qingdao right in front of Haier’s doorstep. So this was a move by Rechi to help their major customer Haier, a move that, besides selling the D-shares later at a profit, was more than compensated by fostering the good business relationship with Haier.
Most IPO-helpers later sold the D-shares back into the market at often only slightly higher prices. This process lasted for a couple of years. The corresponding selling pressure effectively led to a price cap on the D-shares. As the IPO-helpers progressively sold their shares into the market, they were soaked up by funds, institutional and retail investors. Hence, in some metaphorical way, it might be said that Haier did not IPO in a one-time-event in 2018, but experienced an IPO that was spread out over a couple of years.
This metaphorically year-long IPO of the D-shares was still ongoing when Haier created a new share class, H-shares, in December of 2020.
Up until that date, Haier Smart Home owned roughly 51% of publicly traded Haier Electronics. For taking Haier Electronics over completely, Haier Smart Home created H-shares in Hong Kong. Haier Electronics shareholders were treated with courtesy. They got 1.6 Haier Smart Home H-shares for each Haier Electronics share plus a cash compensation. Haier noted that the H-shares could not be converted or exchanged into D- or A-shares.
The mention is important, because Mainland Chinese Shanghai and Hong Kong shares commonly trade at a spread. Because of capital controls, mainland Chinese people focus on investing in the home Chinese stock market, and with this glut of buyers, the traditional premium of A-shares in comparison to H-shares can be explained. Despite the “Stock Connect Program”, which allows qualified investors from China to trade in Hong Kong and vice versa, A- and H-shares with actually the same voting rights and dividend entitlement tend to be listed at different prices. The spread A to H has persisted for decades now and is even reflected in an index:
https://www.hsi.com.hk/eng/indexes/all-indexes/ahpremium
An arbitrage between A- and H-shares — the closing of the spread between them — practically does not take place.
Spread D- and H-shares
While the H-shares were declared not fungible with either A- or D-shares, the H-shares supplied a real reference price for D-shares that the D-shares didn’t have before with the A-shares, because both D- and H-shares can be traded by international investors, while A-shares can not. As of the 23th Feb of 2024, these H-shares trade at 2.79 EUR. Hence, the financial center in Hong Kong currently says that 2.79 EUR per share is the fair value for a Haier share. In arguing what is the proper price for a Haier share, one might prefer to rely on the price on a liquid exchange like Hong Kong rather than an illiquid one like the CEINEX in Frankfurt, especially when analysts see upward potential to the price in Hong Kong.
Upon the creation of the H-share class, many investors saw a golden opportunity. The arbitrage play was conceived as follows: Mainland Chinese had so far never been able to arbitrage with the very cheap D-shares due to strict capital controls. But Hong Kong Chinese were not subject to those controls. So the H-shares should offer Hong Kong Chinese an enticing opportunity to sell the new H-shares and buy D-shares instead, until the price difference between D-shares and H-shares would collapse.
The argument actually made good sense. It must be noted that the stark difference in price between share classes with equal rights, apart from the persistent Chinese Mainland and Hong Kong spread, is rather the exception than the norm. To give an example: The mining conglomerate Rio Tinto has a stock exchange listing in Australia and another one in England. Both types of shares represent the same company value. However, the two share classes are not convertible. What do the two Rio Tinto shares do in far-off Australia and in England? They have been running parallel up and down for years with only a few percent difference between them.
In truth, examples abound, and more exotic ones. Ninety One is a small company from South Africa with a dual listing in Johannesburg and London that trades at roughly the same level on both exchanges. Carnival Corp. has two listings in London and New York, with only a few percent price difference. ABB corporation is listed in Zürich, Switzerland, and on the Nasdaq Nordic exchange in Sweden; here too the price fluctuates around the same level. All these shares are not convertible with each other and the price formation only takes place at the respective exchange.
The argument for the arbitrage play was good, without a doubt. But it didn’t catch up to reality. The wide spread between H- and D-shares simply persisted. Today H-shares are priced at 2.79 EUR and D-Shares at 1.261 EUR, a whopping 121% difference in price. And here’s the really funny part of the story: The H-shares also trade in Germany. In Düsseldorf, München, Stuttgart and Berlin investors buy Haier H-shares for around 2.79 EUR, while they could buy equivalent D-shares in Frankfurt for 1.26 EUR. Hence, we can see how the argument for the arbitrage play was a very rational argument. Given that the H- and D-shares represent the same rights, why would anyone choose to buy the more expensive pieces? But this is the reality we have been seeing and, so far, continue to see.
Tentative explanations for the spread
Chinese companies do not have a good reputation in Germany, as there have been balance sheet manipulations with some minor values listed here. Additionally, there have been cases of financial statement fraud in larger Chinese companies (e.g., Luckin Coffee). Hence, investors may take Haier into collective punishment. In all fairness, though, financial statement fraud can also occur in DAX companies (e.g. Wirecard). And if this was the reason for the wariness of non-Chinese investors, H-shares would be affected to a similar extent as D-shares.
A valid reason for the discount, however, could be liquidity. The D-shares only had a limited free float so far. The majority had been in the hands of the IPO-helpers who either pursued friendly business interests (Rechi Precision) or a higher political interest, including the establishment of the D-shares (China Investment Corp. and others). With roughly 40% of the market cap remaining to this day in the apparently firm hands of HIC and the Silk Road Fund, the market cap in the free float at the 23th of Feb is about 163 Mio. The market cap of the Haier D-shares is at an S-Dax level. For big institutional investors, this is too little to be able to exit quickly and without large price losses if necessary. They thus prefer H-shares if interested. It’s not that funds wouldn’t buy D-shares; currently the 20 top funds hold 49,211,046 of D-shares, led by Fidelity China Special with 9,942,368 shares and followed by Templeton funds; but the amounts they can buy for them are token amounts only that don’t move the needle in their investment results.
Moreover, one might ask what investors get from owning proper shares in a Chinese company. Haier D-shareholders, as proper co-owners of the company, have full voting rights. But what’s the benefit an international investor derives from a general meeting that takes place in China and where absolutely nothing can be decided even with 100% of the D-shares votes cast, because all decisions are made by the major shareholders in the A- and H-shares anyway? Some German investors dryly note that if someone were to offer them an unsponsored GDR in exchange for four D-shares, they would take that offer immediately.
Tentative explanations: The risk of delisting
A specific risk of the D-shares, which does not exist with ADR/GDR, is delisting. The shareholder position in Germany is not among the most protected. German shares can simply be delisted on the basis of a board resolution. Shareholders only have to be paid a compensation at the level of the 6-month average of the share price — which, in the case of Haier D-shares, represents a significant discount to the H-shares. Such a move could be commanded by the Chinese Communist Party (CCP), but to Haier it would offer no advantage. For this, the listing in Germany with less than 3% of the issued shares is too insignificant. An individual shareholder reported from a phone call with IR that a delisting of the D-shares was not in the cards. IR argued that some institutional investors could only acquire this type of share and that a lot of work had been invested in the D listing.
Also, even if a delisting occurred, for investors with a strong conviction about Haier’s future success and an adamant buy-and-hold approach, it would not be a tragedy. Of course, the shares couldn’t be traded any more in the open market, but the payment of the dividends would continue. Only when the payment transactions with China stopped it would look really bleak for D-shareholders. How likely is this latter scenario?
Tentative explanations: The China country risk
An attack by China on Taiwan could certainly lead to the end of tradability of the shares and possibly also to the refusal of dividend payments to foreign investors as part of a tid-for-tad escalation in financial sanctions.
While the CEINEX with its premium listing gives the illusion of the security of the property rights on the Haier company via the D-shares, there are no legally compelling mechanisms to also enforce these property rights against Haier or the Chinese government. In the event that the Chinese government opposes the German shareholders’ property rights, Haier D-share owners would be left empty-handed. Even if they tried to get out in time, there would be no guarantee that there would be enough buyers for their shares. The risk of a sudden total loss is as much present as it was with pre-invasion Russian titles with super low P/E ratios, which became worthless overnight; for foreign investors, it is not possible to receive dividends even from the Russian original shares (let alone the ADRs).
That the same sanctions would be applied as in the case of Russia can be considered unlikely, because of the vested economic interest that the European Union has in China. But it cannot be ruled out that the EU would be willing to “set an example”, even if it would only shoot Western investors in the knee. And amongst its responses to the example-setting, the Chinese government might then rule to prohibit the transfer of dividends abroad.
There is a point, though, that limits the scope of the applicability of the comparison with Russia. Gazprom had a low P/E ratio, a high dividend yield and also a high discount to the valuations of Western oil companies like Shell, BP, Exxon. But Gazprom had practically no new business outside their home market. Haier, to the contrary, operates internationally, with production facilities around the world and roughly half of their revenue being generated ex-China.
Further, Haier is hardly more exposed to the China country risk than, say, Volkswagen. VW derives the bulk of their profits from passenger cars; of the roughly 6.5 million they delivered worldwide in the first nine months of 2023, 2.5 million were sold in China, close to 40%. Imagine those 40% evaporating, then the Volkswagen stock would be toast. BASF is currently betting their future on a Verbund site in Zhanjiang whose construction costs are expected to amount up to 10 billion EUR. To understand the enormity of this investment, it should be kept in mind that at the time of writing this blog post, BASF’s market cap is 44 billion. And albeit Apple’s attempts to diversify its production facilities to other countries like India, recent estimates suggest that 95% of the total iPhone supply still comes from China. All this to say that the Chinese country risk might be just as present, if not more so, in dominant Western companies as in the Chinese company Haier Smart Home.
Still, the argument has been made that Haier Smart Home is at an accrued risk of government intervention, uneconomical action or shareholder disadvantages because 30% of Haier Group, the parent company, is in communal ownership — that is, 30% of the company belongs to its employees. But the argument could actually be reversed: A socialist/communist government will find it much easier to interfere with the rights or wallets of companies that belong only to foreigners or only to individual super-rich Chinese than to interfere with the wallets of Haier employees, who exemplify what might be a goal of the CCP: capitalism for the benefit of the working masses. Despite majority ownership by capitalist investors, in Haier still lives the idea of employee ownership, which represents what might be the politically desired combination of socialist ideology and economic success.
Another consideration has to be taken into account. Undoubtedly, there is a high political risk with Chinese companies. Be it that the CCP nationalizes the company or China attacks Taiwan and is sanctioned or the CCP puts the CEO into jail, etc. However, this risk exists with all shares of Chinese corporations and therefore it is also priced in with all these shares. The H-listing of Haier Smart Home already fully prices in this country-specific risk — whatever it may concretely be. International investors trade on the Hong Kong Stock Exchange and the H-listing is very liquid. No good argument can be made for why a fair market price for Haier shares would not be able to form on the Hong Kong exchange. The H-share also has a discount to the A-share. With the A-share, one could say that the Chinese risk is not seen or priced in. But the H-share and the ADRs traded on the US exchanges based on the H-share are already priced with the China risk discount.
Finally, the China country risk contains various issues, but specifically for the Taiwan risk, analysis suggests that it might not be a question of China deciding or not deciding to invade Taiwan, but of China’s actual ability to invade Taiwan, as the think-tank Council of Foreign Relations has elaborated in the following article: https://www.cfr.org/article/why-china-would-struggle-invade-taiwan?utm_source=substack&utm_medium=email
I’m very much aware that this is a dreadful topic. Sadly, for an investment in any Chinese company this dreadful issue needs to be considered. I very much hope that the Chinese government will work at all times towards securing peace and well-being of all within and outside its borders.
Closing the spread? A: buybacks
As has been mentioned above, investors who scooped up Haier Smart Home D-shares before or in wake of the Hong Kong listing, banking on an arbitrage opportunity to play out, were badly deceived. True, from December 2020 to January 2021 the D-share price surged from ca. 1 EUR to ca. 2.5 EUR, but then went into a progressive downtrend, and has been languishing around the 1 EUR mark for the better part of 2023. Accordingly, this is how the D-share chart looks like (from wallstreetonline.de):
Evidently, in the D-share investing community, the question of how the gap to the H-shares might still be closed is very much present. There are basically two scenarios to be considered.
One of them would be Haier initiating a buyback program for the D-shares. This is not to be confused with Haier International Co. (HIC) buying approx. 20% of the D-shares after the IPO. The buys by HIC were not to the disadvantage of liquidity on the CEINEX, as HIC bought the shares in block trades from the IPO-helpers and not from the free float. HIC buying Haier Smart Home D-shares is not a buyback program. HIC is a subsidiary of Haier Group, just like Haier Smart Home is a subsidiary of Haier Group. So it can be seen as a third party that invests money here, like a private equity fund that acquires stakes.
But what Haier Smart Home does indeed, is buying back H- and A-shares, with its own funds, and cancelling them. This increases the value of the remaining shares in the company. With the D-shares, however, another company of the Haier Group, HIC, bought and has kept them so far. This was not a share buyback program run by Haier Smart Home.
Good Corporate Governance means that one must always look for the most sensible use of capital and if one decides to put a few million into a share buyback program, then it must be conclusively justified why one buys back the more expensive shares rather than the cheapest. In this view, it can be argued that Haier wastes company funds in buying back expensive H- and A-shares. The D-shares are not even considered in the employee participation program, where only A-shares and H-shares figure. Admittedly, apart from buying back shares for the stock based compensation program, the authorisation to buy back H-shares makes sense as it could maintain the share price in turbulent times, and if at some point a capital increase is pending, for example, to finance a larger acquisition, a commendable H-share price is certainly in the interest of all shareholders, because then new shares could be issued at higher prices and less dilution. This having been said, many D-shareholders, excluded from the buyback programs that A- and H-shareholders enjoy, feel treated like third class shareholders.
In 2021 and 2022 Investor Relations (IR) pointed out to individual D-shareholders that they did not want to reduce the free float further with a D-share stock buyback program. It’s true that if Haier wanted to stick to the D-shares, in the long term it is rather further capital increases that would to be expected, because the problem of this share class is the illiquidity. If Haier wanted to celebrate successes with the D-class, the free float market cap would have to rise above 1 billion EUR, then it would also become interesting for big funds to move the needle in their investment results. How Haier’s argument, the preservation of D-share liquidity, squares with HIC’s picking up 20% of D-shares from the IPO-helpers instead of letting those 20% increase the free-float, is a legitimate question, but probably just reflects Haier trying to find the sweet spot between supporting the D-share price and preserving the free float at the same time.
So D-share buybacks were generally not considered.
But in 2023, the tune changed and IR answered to surprised shareholders that they have actually been trying to open a trading account for a year already. According to one shareholder, this is the concrete answer he got: “We have been trying to open an equity account on behalf of Haier Smart Home with one of the banks in Germany since December 2022, without which the buyback could not be launched.” And: “A trading account is required to carry out activities such as share buybacks in Germany. We have provided all the documents to the Bank, however, we have not yet been informed of the account opening progress.” Besides, Helmut Fink of Verus Capital reported having spoken to representatives of Haier at the Equity Forum (EKF), where they told him that Haier was undertaking steps towards a buyback of D-shares.
To this creation of expectancy of a buyback, however, it might be responded that ‘We are in the process of opening a trading account’ could be the Asian-polite for an unabashed ‘No’. After all, it is somewhat difficult to fathom how Haier, which has operating businesses in Europe, has been finding itself unable to open a trading account for over a year now.
From a shareholder perspective, a buyback of D-shares and subsequent cancellation of these shares would naturally make short-term sense (as then a higher proportion of the company remains per share received), but would only make sense if the rest of the shares was then converted to H-shares, as the liquidity of the D-shares would otherwise decrease further and the whole D-share program would make even less sense.
Closing the spread? B: Conversion to H-shares
The company statutes provide for the possibility of a conversion of D- to H-shares. The statutes provide for the possibility, not the actuality. So this is theory, as there would be an application process and both Haier’s Annual General Meeting and various authorities would need to approve the move. Individual shareholders relate from their contact to IR in the past years that Haier wanted to check the convertibility between D- and H-shares, but ultimately shied away from it because it would have required the approval of the German, Chinese, and Hong Kong authorities.
Arguably, even if the D-shares are of little value for Haier after the H-listing, the Chinese authorities do not yet know whether they want to further develop the D-share market. For a future of CEINEX it would of course be detrimental to remove the only D-shares that exist after just a few years by converting them to H-shares. If a conversion was now made possible, lots of investors would convert immediately and sell in Hong Kong for a quick profit, and the whole D-share program would have been for the cat. Similarly, if the D-shares were converted into GDRs, it would be equivalent to a conversion to H-shares, as GDRs are depositary receipts thereof.
Still, despite the bureaucratic hassle and the political component, there might be a chance that D-shares will eventually be converted into H-shares. And the argument behind this remote chance is that for Haier, the D-share class, as said above, no longer makes sense. They have successful international shares, the H-shares, and the D-shares are languishing with a wide discount without a positive marketing effect. A future issuance of D-shares is highly unlikely. Why should Haier force a capital increase at low prices on a market that has already completely disappointed expectations at the IPO and afterward, when it has the liquid H-shares which all international investors can buy anyway? It is clear that Haier will not issue such cheap D-shares again as it did for the financing of the acquisition of Candy, as Hong Kong Dollars from the placement of H-shares can also be freely converted into Euros and other currencies.
Arguably, though, Haier would not even need to do a D- to H-shares conversion, but only make the path for a conversion official and recognisable: ‘look here, that’s the manual, that’s the official way to convert your shares, step by step’. The price would quickly adjust and market participants would not necessarily want to convert anymore. The conversion is only really desirable as long as there is the big price difference. If the difference is gone, everyone will like D-shares just as much as H-shares. In this way, Haier could continue its three-platform capital market strategy.
Finally, in the event that the company were to delist the D-shares, which has been contemplated above, management could give no conclusive reason against a conversion to H-shares other than retaliatory action against foreign investors. Once the D-shares delisted, no argument could be put forth for how a conversion could run counter to the interests of D-shareholders.
The toe in the pond
In summary, if Haier truly wanted to reduce the price spread between D- and H-shares, they would have options. Starting with free convertibility, buybacks of the D-shares instead of the expensive H- or A-shares, and also, more active IR work.
Haier is not a victim of circumstances here, instead, they have consciously decided to just dip a small toe in the pond. The D-shares trade at a discount because Haier lets them trade at a discount.
How the story of the spread between the share classes will play out in the next few years will be interesting to watch.