MSCI ACWI ex-USA index underweights China and India. Is that a governance discount? Should investors push MSCI to aggressively weed out poorly governed stocks in these countries instead? Indian policy makers may want to look at ways to overcome the “India discount.”
China and India flag together realtions textile cloth fabric texture
Increasingly, retail investors and even institutions rely on pre-packaged indexes to build wealth for the future. Perhaps only the most sophisticated of these customers look under the hood to critically evaluate what is under the wrapper of the index. This raises questions about the index maker’s fiduciary responsibility related to constructing and maintaining that index.
Consider the case of the well followed international index, MSCI ACWI ex-USA. Investors use products reliant on this index to get exposure to international stocks. MSCI’s tear sheet states, “The MSCI ACWI ex USA Index captures large and mid-cap representation across 22 of 23 Developed Markets (DM) countries (excluding the US) and 24 Emerging Markets (EM) countries. With 2,321 constituents, the index covers approximately 85% of the global equity opportunity set outside the US.”
Governance discount via country weights in MSCI
Without looking at the data, I would have assumed that an international investor seeking exposure to overseas economies would want to hold securities that broadly reflect the GDP of major countries. Outside of the US (GDP of $27 trillion), the top 10 economies, sorted on GDP are: (i) China with a GDP of $17.7 trillion; (ii) Germany with a GDP of $4.4 trillion; (iii) Japan with a GDP of $4.2 trillion; (iv) India with $3.7 trillion; (v) UK with $3.3 trillion; (vi) France with $3 trillion; (vii) Italy with $2.1 trillion; (viii) Brazil with $2.1 trillion; (ix) Canada with $2.1 trillion; and (x) Russia with $1.8 trillion. This quick and dirty analysis, of course, assumes that our investible universe is reduced to 10 countries outside the US, for simplicity.
These proportions would dictate the following country weights in MSCI’s ACWI ex-USA index: 40% for China, 10% to Germany, 9% to Japan, 8% to India, 7% each to UK and France, 5% each to the remaining countries (Italy, Brazil, Russia).
You could, of course, argue that GDP per se is not an investible security, and one should ideally look at the proportion of stock market capitalizations by country. On top of that, GDP represents the value add of sectors other than the publicly listed domain, which is the only segment open to MSCI’s equity investments. However, stock market capitalization can pick up astute overseas investments of a country’s companies (e.g., Japan) although the country’s GDP may not grow rapidly.
The top 10 countries, by proportion of market capitalization, beside the US (market cap valued at $49 trillion), are (i) 27% for China ($10.6 trillion); (ii) 14% for Japan ($5.4 trillion); (iii) 11% for India ($4.2 trillion); (iv) 10% for Hongkong ($3.9 trillion); (v) 7% for France ($2.8 trillion); (vi) 7% for UK ($2.8 trillion); (vii) 7% for Canada ($2.6 trillion); (viii) 2.4% for Saudi Arabia ($2.4 trillion); (ix) 2.2% for Germany ($2.2 trillion); and (x) 2.4% for South Korea ($2.1 trillion).
The actual country weights assigned by MSCI ex-ACWI are, in contrast, as follows: 15% to Japan, 10% to UK, 8% to China, 8% to France, 8% to Canada, 1.46% to Brazil, 1.69% to Italy and 1.09% to Hongkong.
So, one could conclude that China and India are under-represented. China is severely under-represented (9.4% on MSCI including Hongkong although China’s GDP is 37% of our 10-country aggregate and 26% of the 10-country market capitalization). India is modestly under-represented (4.68% on MSCI relative to 8% of GDP and 10% of market capitalization).
In contrast, UK and Canada are over-represented. MSCI assigns 9.58% of its portfolio to the UK although that country only accounts for 7% of the top 10 country aggregate GDP and market capitalization. Canada gets an allocation of 7.55% in MSCI’s portfolio although it represents 4% of the top 10 country GDP and 6% of the top 10 country aggregate of market capitalization.
I have not analyzed other constraints that MSCI faces such as headroom for foreign investors to invest in in specific companies. Subject to that statement, can one interpret this tilt as the ultimate governance discount? India and China, relative to Canada and UK, are not seen as countries that are friendly to outside shareholders. Or is it one more case of home bias - investing in securities closer to home as opposed to markets that feel foreign and difficult to understand?
To be fair, MSCI does market a GDP adjusted version of the ACWI index. This version over-corrects for the governance problem in that Tencent, the Chinese company, has a higher weight in the index than Apple. Surely, Apple is better governed than Tencent. Moreover, I could not find many ETFs that rely on the GDP adjusted version of the ACWI and hence could not verify the exact nature of the adjustment applied.
Understanding one country in detail: India
The MSCI ACWI ex-USA index, as per the holdings published by BlackRock’s ishares version of the ETF based on that index on December 23, 2023, devotes roughly 4.68% of its $4.6 billion of AUM (assets under management) to Indian equities and carries 132 Indian stocks and involves a total mark to market value of around $215 million.
How did MSCI decide which 132 Indian stocks to include? Recall MSCI’s mandate of covering 85% of the world market by value. It turns out that 85% of India’s market capitalization of $4.2 trillion is roughly $3.6 trillion. I suspect MSCI follows a simple heuristic of finding the top N firms, sorted by market capitalization, and cumulatively adding up to around $3.6 trillion at the time of balancing the index, subject to the constraints associated with liquidity, allowing access of firms to foreigners and the sheer number of Indian securities that MSCI wants to hold.
To verify this conjecture, I downloaded the list of the top 200 Indian companies, sorted by market capitalization. It turns out that the cumulative market capitalization covered by the top 132 companies on the list of the top 200 companies, sorted by market capitalization as of 12/30/2023, is approximately $3.1 trillion. This seems reasonably close to the heuristic I outlined.
Exceptions to the heuristic
I cross-tallied stocks that show up in the top 200 stocks, sorted on market capitalization, with the list of India holdings in ishares documents. The idea was to check whether MSCI mechanically followed a rule of picking stocks that added up to 85% of stock market capitalization or were there deviations. To place the data in context, note that the Indian firm with the smallest weight in MSCI ACWI ex-USA India portfolio was Muthoot Finance with a market cap of around $7 billion. I have tried to ignore stocks that were in the top 200 list but had market caps of less than $7 billion to avoid noise caused by volatile stock market valuations of specific firms.
The following companies, that would otherwise qualify, by virtue of market capitalization, did not show up in MSCI’s India portfolio in its ACWI ex-USA index:
(i) Life Insurance Corporation (LIC) with a market cap of $63 billion
(ii) Adani Energy with a market cap of around $29 billion.
(iii) Hindustan Zinc with a market cap of $16 billion
(iv) Indian Railway Finance with a market cap of $15.6 billion
(v) Rural Electrification with a market cap of $13 billion
(vi) Adani Total Gas with a market cap of around $13 billion
(vii) Punjab National Bank with a market cap of around $12.6 billion
(viii) Adani Transmission with a market cap of $12 billion
(ix) Union Bank with a market cap of $11 billion
(x) Indian Overseas Bank with a market cap of $11 billion
(xi) Mankind Pharma with a market cap of $9.5 billion
(xii) Canara Bank with a market cap of $9.4 billion
(xiii) Vodafone Idea with a market cap of $9.4 billion
(xiv) Shriram Transport with a market cap of $9.3 billion
(xv) IDBI (Industrial Development Bank of India) with a market cap of $8.72 billion
(xvi) Zydus Life Sciences with a market cap of $8.4 billion
(xvii) Bosch India with a market cap of $7.84 billion
(xviii) NHPC with a market cap of $7.74 billion
(xix) Alkem Labs with a market cap of $7.5 billion
(xx) National Mineral Development Corporation of $7.3 billion.
Now, let’s ignore the stocks below $10 billion to avoid routine rebalancing type issues that occur when stock prices fluctuate. Moreover, MSCI might want to avoid unnecessary turnover in its index to keep transaction costs and the associated expense ratios lower. The question that now stands is why did MSCI exclude LIC, Hindustan Zinc, Indian Railway Finance, Rural Electrification, and the two Adani companies and the three-state owned Indian banks?
The first patten that emerges is that most of the excluded companies are predominantly state owned (LIC, Hindustan Zinc, Rural Electrification, Indian Railway Finance, and the three-state owned Indian banks). LIC is not in the index on account of low free float. It appears as though Rural Electrification has entered the MSCI India index and might eventually enter the MSCI ACWI ex-USA. I came across an announcement that Adani Gas and Adani Transmission were deleted from MSCI Index in May 2023 because these stocks traded at “their lower price limits for more than 5 minutes cumulatively.” Note that four other Adani group stocks are in the current index. So, the exclusion of Adani Gas and Adani Transmission seems to be driven by technical reasons and not by the red flags raised by Hindenburg, the short seller.
More important, for the purposes of this piece, none of the exclusions appear to be driven by governance considerations. Hence, MSCI’s model appears to be broadly to pick the top N firms by market cap and wait till some public scandal emerges. Shorting of stock is particularly difficult in India. Hence, public revelation of bad news is systematically delayed. In markets where shorting is difficult, MSCI want to consider proactive due diligence to weed out poorly governed firms. More important, Indian policy makers might want to work harder to overcome the perception of an “Indian discount” by making it easier for naysayers to short stocks in domestic markets.
To be fair, MSCI does offer a China A Inclusion Top 200 Select Governance Quality ex Controversies Score Index (CNY). The fact sheet states that “the financial aspects of the Quality factor, measured by the fundamental data – high return on equity, low financial leverage, and low earnings variability, as defined in the MSCI Quality Index methodology, are discounted using several measures of corporate governance.”
I could not find details on how governance was measured but the MSCI World Governance-Quality Index factsheet says it relies on “independence and diversity of board of directors, ownership and control structure of the company, accounting practices and auditor opinions.” Setting aside the merits of these measures, I wonder why this is a special product and not the default version. Another issue is that these indexes conflate fundamental quality of the stocks (as in high ROE etc.) with governance quality. Hence, it becomes hard to isolate the impact of governance alone.
Moreover, I could not find many ETFs built on these governance quality indexes. Hence, I could not verify how exactly this works in practice as MSCI does not publish these weights and holdings for the average Joe to access. Does the absence of governance quality adjusted ETFS reflect lack of investor demand or investor ignorance?
Overseas listing as a signal of higher governance
Coming back to India, another proxy for a firm willing to sign up to higher governance standards is cross-listing on overseas markets. The list of such Indian companies is quite modest. ADRs or American Deposit Receipts are available for the following Indian firms on US exchanges: HDFC, Dr. Reddy’s, ICICI, Infosys, Wipro, Eros International, Wipro, HDFC, Genpact, Tata Motors, Vedanta, Videocon, WNS Holdings, MakeMyTrip, Dr. Reddy’s Lab, Azure Power, iGate, and Axis Bank.
Ideally these firms should get a higher allocation weight in the MSCI ACWI ex-USA index, if they qualify otherwise to be in the index.
What does MSCI ACWI do when governance concerns are publicly raised?
MSCI appears to have cut its weights assigned to the four Adani firms in the index. The announcement was couched as a technicality: “it reassessed the size of companies’ free floats and determined that there was “sufficient uncertainty” surrounding some investors in Adani companies.” But it appears as though MSCI’s response is purely reactive. It would have been highly unusual if MSCI had not acted given the global storm that erupted from the Hindenburg report related to the Adani Group.
Billions of dollars of investments in India follow MSCI’s lead. The meta question for these investors: are you better off with severe underweights in China and India in MSCI ACWI ex-USA index? How useful would it be to increase the weights to these countries but aggressively weed out poorly governed stocks in these countries instead, as a matter of course? Would the incremental cost of following a targeted strategy of weeding out poorly governed firms be the worth the potential subsequent tail risk or even the perception that does your institution does not care all that much about the G in ESG?
