Breaking the LOOP
A fundamental rule in finance is the Law of One Price (hereinafter LOOP), stating the price of identical goods must be the same across markets. Formally: “...an identical asset or commodity will have the same price globally, regardless of location” (Investopedia, 2020). A corollary of the LOOP is the law of no-arbitrage. In finance, an arbitrage is the “simultaneous purchase and sale of the same asset in different markets to profit from tiny differences in the asset’s listed price” (Investopedia, 2021). The law of no-arbitrage is widely used to price assets such as derivatives. For example, a forward rate agreement (FRA) should “lock in” a rate such that it is not possible to short (long) the FRA and long (short) the spot rate and make a risk-free profit. Formally:
with ft(T1,T2) as the forward rate from T1 to T2, and rt,T2 and rt,T2 as the spot rates from t to T2 and t to T1 respectively. In a similar fashion, if one purchases gold in Paris, Shanghai, New York, or Dubai (accounting for currency exchange) the price should be the same across markets. Likewise, a stock from a dual-listed company should, under the Law of One Price, have a similar price in the different stock markets where it is available. For this rule to hold, there must be an efficient, frictionless market, that is, a market with negligible transaction and transportation costs, a sufficient demand and supply of the assets as well as an absence of legal restrictions (Investopedia, 2019). Should any difference in the price of an asset in different locations be observed, arbitrageurs and investors worldwide would quickly seek to benefit from the opportunity to make a risk-free profit.. The massive purchase of the asset in the location where it is under-priced, and massive sale where it is overpriced, would lead to the creation of a new, cross-location equilibrium price. While in an efficient market, arbitrage opportunities do not exist, they regularly appear in our world but are quickly corrected by market forces (Dothan, 2008).
For over twenty years, however, the Chinese stock markets appear to have violated this golden principle, and dual-listed companies on the mainland and Hong Kong exchanges sell at sizable premiums in the A-share market. This constitutes a clear violation of the LOOP, but not an arbitrage opportunity, as otherwise supply and demand would have corrected the deviation.
This points to the existence of some friction or location-specific factor which can explain the persistence of the violation. There must also exist time and company-specific factors which explain the time and company variations in premium size.
Market overview
China has three stock markets: two onshore in Shenzhen (Shenzhen Stock Exchange, SZSE) and Shanghai (Shanghai Stock Exchange, SSE), and one offshore in Hong Kong (Hong Kong Exchange, HKEX). Whilst roughly equal in terms of listed companies, the Shanghai exchange clearly dominates in terms of market capitalisation, meaning that companies listed in Shenzhen, and in particular in Hong Kong, tend to have a larger market capitalisation (and are thus likely larger in size).
FTSE Russel (2019) offers the following categorisation of stocks in China. The A-shares and B-shares markets refer to stocks of mainland Chinese companies traded in one of the two exchanges situated in the mainland. A-shares are quoted in the local currency, the renminbi (RMB), whereas B-shares are traded on the same exchanges but traded in a foreign currency. Both A and B shares are incorporated within the PRC. N-shares are shares of mainland companies listed in a New York stock market (NYSE, NASDAQ, or NYSE MKT), which have been incorporated in a country other than the mainland and are traded in USD. Finally, H-shares are traded in Hong Kong, and in HKD, but are from companies incorporated within the PRC. Other share classes exist, such as P Chip or S Chip, but those are of little relevance here.
As of May 2023, there were 4555 companies issuing A-shares, 38% of those listed in Shanghai and 62% in Shenzhen. In comparison, at this date, there were only 305 H-shares (HKEX, 2023) issuing companies and just 148 dual-listed ones. The market capitalisation of the H-shares market was only HKD 5.924tn (HKEX, 2023) on the same date.
The Shanghai and Shenzhen stock exchanges were first opened in 1991 as part of Deng Xiaoping’s (Coase & Wang, 2011) reforms to modernise the Chinese economy, and soon after, in 1993, Tsingtao Brew was listed in both the SEHK and SSE (FTSE Russel, 2017), effectively becoming the first A+H company. Their numbers grew steadily, especially after 2005, to reach 148. An overwhelming majority (118) of those stocks are dual-listed in Shanghai and Hong Kong, as opposed to 30 dual-listed for Shenzhen and Hong Kong.
The A-shares market has a set of relatively unique characteristics which the H-shares market does not exhibit. First, it is relatively segmented from the rest of the world, and capital movements in and out of the mainland are strictly regulated (Xiao & Kimball, 2004). Second, almost all investors are mainlanders, and 80% of those are retail investors (Calhoun, 2020). This is highly unusual, as most markets would have a more significant share of institutional investors (Calhoun, 2020). Lastly, it has been dubbed a “casino” (Amadeo, Kimberly, & Boyle, 2020), and even if the situation has improved, the lack of transparency, the vast proportion of retail investors, and a large number of small stocks do make the A-shares market relatively more volatile than other major exchanges.
The AH premium
Under the LOOP, those A- and H-shares of the same company should theoretically be sold at the same price in all the markets in which they are listed. This is however not the case, and violations of the LOOP are not only significant in scale but persistent in time. This is a well-known phenomenon and the focus of some of the current financial literature.
Formally, we can define the premium for company at time as:
Those premia are overwhelmingly smaller than 1, meaning that H shares sell at a discount, and conversely, A shares sell at a premium. Only one company (CM Bank), out of 148, is currently more valuable on the Hong Kong exchange than on the mainland’s. Meanwhile, 64 entities trade at more than a 50% discount in Hong Kong, and some close to 90%. In other terms, buying Holly Futures in the Hong Kong market is 90% cheaper than in Mainland China. This is not just a violation of LOOP, but rather it is an astonishingly gargantuan and systematic one. As Figures 1 and 2 show, were it not for hidden hindrances, there would be numerous sizeable “free lunches” for hungry traders to munch on.
The Hang Seng Stock Connect China AH Premium Index (HS CAHPI) tracks the evolution of the premium.
With as the Price of j-shares of company at time, as the total issued j-class shares of company , as the free float-adjusted factor of j-class shares of company (i.e., stocks that are available for public trading, thus excluding restricted shares, which is a very significant factor in the mainland and excludes about 66% of A-shares), and as the foreign exchange rate of a traded currency of j-class shares vs USD at time . This is, therefore, a weighted average index depending on the free float-adjusted market capitalisation of companies (Hang Seng, 2017), which provides a useful indicator of the evolution of the premium as a whole.
The historical average of the premium stands at 123.54, with spikes close to 200. The index is currently valued just shy of 142, indicating that A shares trade at a 42% premium on average This means that purchasing a share from an A+H company in one of the mainland Chinese exchanges is on average 42% more expensive than in Hong Kong. Figure 3 reveals the persistence and magnitude of the violation, as well as its substantial volatility.
Figure 3: The evolution of the premium over time. Any number above 100 indicates A-shares sell at a premium.
For most of its existence, this premium was positive, meaning that A-shares were more expensive than H-shares. According to the laws of supply and demand, this would indicate that there is a larger demand or a lower supply of A-shares. Besides, the premium's existence and persistence reveal serious frictions concerning capital flows in and out of the mainland.
This last point is especially concerning. While it is widely documented (Xiao & Kimball, 2004; Ma & McCauley, 2008) that there exist barriers to the free flow of capital between the mainland and the rest of the world (in both ways), those barriers are gradually being reduced. A good example of such reductions is the implementation of the Stock Connect schemes. Indeed, the Shanghai-Hong Kong Stock Connect was implemented in late 2014, which coincides with a spike in the premium. Basic intuition, however, would dictate that the relaxation of restrictions on capital flow and increased connectivity should have smoothed the price difference. This seems to indicate that the increase in the demand for H-shares by mainland investors was greatly outweighed by that in A-shares by the rest of the world. The Shanghai-Hong Kong Stock Connect scheme was launched on November 17th, 2014, and a similar Shenzhen-Hong Kong scheme was launched on the 5th of December 2016. Both schemes work in the same fashion and allow some investors (with no less than 500 000 RMB in their account) in each exchange to trade shares in the other market using local infrastructures (brokers, clearinghouses, etc.). Only secondary market trading is available on the platform, for a number of eligible stocks and subject to a daily quota: 52bn RMB Northbound (from Hong Kong to the mainland) and 42bn RMB Southbound (from the Mainland to Hong Kong) for each scheme in Shanghai and Shenzhen (HKEX, 2021).
Another explanation for the premium is that of foreign ownership. Hong Kong has traditionally been the gateway to China, and its stock exchange is vastly more internationalised than A-shares markets. The theory would suggest that a market with more international investors would have a higher valuation. However, the effect is once again the opposite of what theory predicts: the less international market has a higher valuation. While the explanation of foreign ownership is unsatisfactory, it does lead to the question of the demographics of investors. Hong Kong attracts a larger amount of foreign capital and institutional investors, while the A-shares market is the realm of retail investors (Calhoun, 2020).
A corollary to the explanation of investor demographics is the price elasticity of demand. Since the Mainland stock exchanges are dominated by retail investors who have little choice to invest in other assets, their demand is probably less price elastic, leading to a price premium.
Finally, the quality of financial regulations may play a role in the existence of the premium. Past research found that stronger regulation led to a higher valuation (Balani, 2019). This is visible, for example, with the N-shares, where Chinese companies listed in New York receive a higher valuation. This is the so-called “cross-listing premium”. However, the SEHK has overall more stringent, more extensive regulations. Its accounting practices uphold widely recognised international standards, and the overall environment is much more transparent than that of the mainland. Again, this theory falls short of explaining the existence of the premium since H-shares sell at a discount which is the reverse of the theory's prediction.
Besides failing to explain the existence of the premium, those explanations do not account for the variation across time and companies. Indeed, the premium is not evenly distributed across the 148 companies. It is, therefore, paramount to our understanding of the premium to investigate the time and cross-sectional variations of the premium.
Determinants of the premium
A number of factors have been used to answer the triple question of why this premium exists, why it varies across companies, and why it varies in time.
Why is there a premium? The existence of a large and persistent premium is puzzling, although several causes have been identified.
- Capital account inconvertibility. This is by far the largest reason behind the existence of the premia. Were the RMB to become freely convertible and the PRC to relax its stringent cross-border flow restrictions, the AH premium would undoubtedly be relegated from anomaly to past curiosity.
- Information asymmetries. Obviously, the more information asymmetries there are, the lower the premium will be. Exchanges in the mainland not only suffer from a language barrier, but they also have lower reporting, disclosure, and governance standards (Zhang et al., 2020).
- Investor demographics. As aforementioned, the A-share market is dominated by retail investors with speculative incentives and no alternative investments. This is evidenced by “its high turnover ratio and short investment terms” (Zhang et al., 2020). Overseas and institutional investors account for “3% and 20% respectively” of A-shares investors, against 46% and 60% in the H-share market.
- Market infrastructures. The Hong Kong market is overall much more flexible. For example, settlements in Hong Kong are done at T+0 against T+1 in the mainland, net short selling is allowed in Hong Kong, and both the SSE and SZSE have more stringent “circuit breakers” (e.g. 15 minutes suspension after a 5% movement in any direction, and whole day recess following a 7% change against a 5 minutes cool off after a 10% variation, CNBC, 2020).
- Cross-sectional variations. The premium varies in size across companies.
- Company size. All else equal, a larger company in market capitalisation has a lower premium. This is explained via two channels: larger companies uphold higher reporting standards (lower information asymmetries, as evidenced by Xuechun et al. 2020), and the fact that size is positively correlated with liquidity.
- Liquidity. As evidenced by Amihud (2002), illiquidity yields a premium. Illiquid A-shares are thus priced higher, increasing the premium.
- Dividend ratios. All else equals, stocks with larger dividend payments are more attractive, including for foreign investors (for tax-related reasons). A higher dividend pay-out thus reduces the premium.
- Time series variations. The premium fluctuates over time.
- Foreign exchange rates. The HKD is tied to the world’s reserve currency by a crawling peg. Arquette et al. (2008) find that changes in foreign exchange rates, and foreign exchange expectations can explain 40% of the premium’s size (ceteri paribus, a stronger USD increases the size of the premium). In particular, Zhang et al. (2020) find that the dollar’s strength throughout 2015 coupled with poor expectations about the Chinese economy explain the sharp increase of the AH premium after the introduction of the connect schemes.
- Market sentiment. Unsurprisingly, buoyant expectations about China’s stock market increase the size of the premium, as demand for A-shares increases both because of the increase in wealth in the mainland, the relocation of domestic investors’ investments towards the domestic stock market, and thanks to foreign investments.
- Stock connect schemes. Opened in 2014, the stock connect schemes have significantly improved the cross-border financial flows (Xuechun et al. 2020). Although no analysis has yet been released on the swap connect scheme, we expect it to have a similarly negative effect on the size of the premium.
Such analysis is precisely why the AH premium is fascinating: it reveals arcane market mechanisms and behaviours, exotic effects of capital controls, and unforeseen consequences about financial fundamentals.
About the Author
Nicolas Guignard is a master student candidate within the Economics and Public Policy
track of Sciences Po’s school of Public Affairs. He holds two bachelor’s degrees from
Sciences Po and the University of Hong Kong (in Economics and Finance). His experiences nurtured his long-time interest in China’s development framework. His main area of expertise revolves around economics and economic policymaking. He currently works for Amundi Asset Management as a macroeconomic researcher. You can find him on LinkedIn. https://www.linkedin.com/in/nicolas-guignard-7790b71a0/
The opinions expressed here are those of the writers and do not represent the views of European Guanxi.
Do you have an article you would like to share? Write for us.
Bibliography
aastocks. (2023, June 16). A+H. A+H - AH Shares.
AFM Editorial Office. “China A-Shares: All Investors Need to Know .” AsiaFundManagers.com, AFM Editorial Office, 28 Sept. 2020.
A/H Shares Premium｜Top Picks｜SG Warrants & CBBCs, Société Générale.
Amadeo, Kimberly, and Michael J. Boyle. “Why China's Stock Market Is Like a Casino.” The Balance, 21 Nov. 2020.
Amihud, Y. (2002) Illiquidity and stock returns: Cross-section and time-series effects. Journal of Financial Markets 5 (1): 31–56
“An Overview of A+H Shares and AH Discounts and Premiums.” Fund, Fund, Dec. 2020.
Arquette, G. C., Brown Jr, W. O., and Burdekin, R. C. (2008) US ADR and Hong Kong H-share discounts of Shanghai-listed firms. Journal of Banking & Finance 32(9), 1916–1927.
Asness, Clifford S. “The Power of Past Stock Returns to Explain Future Stock Returns.” SSRN, 8 Nov. 2016.
Balani, Henry. “Assessing the Introduction of Anti-Money Laundering Regulations on Bank Stock Valuation: An Empirical Analysis.” Journal of Money Laundering Control, Emerald Publishing Limited, 7 Jan. 2019.
Calhoun, George. “A Strange New Bubble In Chinese A-Shares (A Is For Arbitrage).” Forbes, Forbes Magazine, 30 Aug. 2020.
Capturing the Chinese A-Shares and H-Shares Anomaly. FTSE Russel, 2016.
Carpenter, Jennifer N., et al. “The A-H Premium and Implications for Global Investing in Chinese Stocks.” 6 Feb. 2020.
“China A-Shares - What Are They and Can You Buy Them? Find out the Answers to These and 27 Other Essential Questions Here.” Asset Management - Global, UBS, 6 Nov. 2020.
Coase, Ronald, and Ning Wang. How China Became Capitalist. Palgrave Macmillan, 2011.
Dothan, Michael U. Efficiency and Arbitrage in Financial Markets. Research Gate, Jan. 2008.
Downey, Lucas. “Frictionless Market .” Investopedia, Investopedia, 16 Sept. 2020, www.investopedia.com/terms/f/frictionlessmarket.asp.
Fernando, Jason. “What Is Arbitrage?” Investopedia, Investopedia, 7 May 2021. www.investopedia.com/terms/a/arbitrage.asp#:~:text=What%20Is%20Arbitrage%3F,markets%20or%20in%20different%20forms.
“For the SAME COMPANY, China A-Shares Currently Trade at a 48% Premium to H-Shares. How to Invest?” The InvestQuest, 15 Oct. 2020.
Guide to Chinese Share Classes - FTSE Russell. FTSE Russel, Dec. 2019.
.
Guo, Lin. Examining the Price Difference for Chinese A, H and N Shares. Aug. 2004.
Hong Kong and Mainland Market Highlights, HKEX, 5 June 2023.
“Index Methodology for Managing Hang Seng Stock Connect China AH Index Series.” Hang Seng, Hang Seng, Sept. 2017.
“IShares MSCI World ETF: URTH.” BlackRock, BlackRock, 2021.
“Law of One Price.” Investopedia, Investopedia, 7 May 2021.
Lee, Yen Nee. “‘Circuit Breakers’ Are Being Triggered across Asia — Here’s How They Differ across Major Markets.” CNBC, 13 Mar. 2020.
Ma, Guonan, and Robert N McCauley. “Efficacy Of China'S Capital Controls: Evidence From Price And Flow Dat.” Pacific Economic Review, Wiley Blackwell, 9 Jan. 2008.
Stock Connect Another Milestone. HKEX, 4 Feb. 2021.
Xiao, Fengjuan, and Donald Kimball. Effectiveness and Effects of China's Capital Controls. 2004.
Xuechun, Zhang, et al. AH and A-ADR Premiums. The People’s Bank of China, 2 Nov. 2020.
Zhang, Renbin, and Tongbin Zhang. “AH Premium: A Natural Experiment.” SSRN, 12 Nov. 2020.
Zhang, Tongbin, and Renbin Zhang. “Understanding AH Premium in China Stock Market.” Understanding AH Premium in China Stock Market-|Antai Business College Shanghai Jiaotong University, 22 Nov. 2016.
Comentarios