Japan fund flows part 1: who’s been buying?

by SR on September 30, 2012

Tokyo Stock Exchange combined traded value by investor type in 2011, in trillions of yen

Who buys Japan? And who sells?

In previous posts (such as this and this) I have considered sector performance in the Japanese market, specifically the relative performance of the 33 industry sub-indices of the TOPIX market index. My focus has been autos, auto parts and machinery, because those are areas in which I have experience. I also covered technology stocks for many years, but technology in Japan is a tough row to hoe, as the implosion of once-respected industry giant Sharp shows.

In today’s post I’m going to pull back from the nitty-gritty detail of sector-specific analysis and explore which investors have been buying Japan and which have been selling, at least in 2012. In future posts we’ll consider investor buying and selling and see if we can link either to particular stages of bull (and bear) markets over the past decade. In this post we’ll just have a look at market participation by different investors as revealed by flow-of-funds data.

The conclusion is one you have no doubt already guessed: foreigners dominate the market, were big buyers in the first three months of the year and have been big sellers thereafter. While trust banks and other domestic buyers have stepped into the gap from April 2012 onward, they haven’t been able to do more than soften the impact of foreign selling. This quantitative view fits well with what I hear from and about foreign investors, which is that they are basically frozen in the headlights. They don’t want to buy, or sell, or read research until they have some clarity on the future of Europe and everything affected by it (pretty much everything). What can you do in such a market? Avoid speculative stocks like the plague; do not assume bad news is discounted in advance; don’t sell your winners. Toyota and Denso would be beneficiaries of such tactics as they are staid companies, enjoying a very good year and have already performed well in 2012.

Where the data comes from, what it means

The Tokyo Stock Exchange (TSE) releases weekly and monthly data on buying and selling by different classes of investors. This data is freely available on the TSE’s web site; the difficulty lies in collating, organising and pulling it all together into something coherent. The survey is not all-inclusive – it covers only trading participants with capital of more than 3 billion yen – but seems to be pretty comprehensive. The part of the survey that we are using covers the first and second sections for the Tokyo, Nagoya and Osaka stock exchanges. In 2011, this section of the survey covered 93.1% of the total traded value. The figures that we are using in this post do not include the Mothers exchange, but as Mothers only accounted for 1.3% of the traded value of the 1st and 2nd sections of the Tokyo, Nagoya and Osaka exchanges in 2011 I don’t feel that the exclusion of Mothers is an issue.

Tokyo Stock Exchange traded value by investor type, as percent total traded value

Each month and week, the TSE releases spreadsheets of the buying and selling data. Each spreadsheet contains data on ‘sales’, ‘purchases’, ‘balance’, which is the the net of sales and purchases, and ‘total’, which is the combination of sales and purchases. Each spreadsheet reports a ‘Total’ category, which is broken down into ‘Proprietary’ and ‘Brokerage’. The Brokerage category is split further into ‘Institutions’, ‘Individuals’, ‘Foreigners’ and ‘Securities companies’. Institutions is further broken out into ‘Investment trusts’, ‘Business companies’, ‘Others’ and ‘Financial institutuions’. Finally the Financial institutions category is sub-divided into ‘Life and non-life insurers’, ‘City and regional banks’, ‘Trust banks’ and ‘Other financial institutions’. (See the bottom of this page for more on definitions.)

The chart at the top of the page shows this organisation. The figures used in that chart are the combined sales and purchases for each category; it is not the net figure. The net figure is also important and we will look at it later, but to get a grasp of who is paying the commissions in Japan’s stock markets, we should look at both buying and selling. And as it happens, foreigners accounted for 359.3 trillion yen out of a total brokerage figure (as per the survey) of 547 trillion yen in value traded in 2011. That’s equivalent to 65% of the brokerage category, a number that is often bandied about when sell-side firms discuss their research firms and on which class of investors they should be focusing. Total traded value in 2011 was 675.6 trillion yen, which includes 128.6 trillion in proprietary trading. Unless there is a reason to assume that foreign participants are much less active in proprietary trading, the ratio of foreigners in proprietary should be similar to that of the market as a whole.

A couple of paragraphs above we have a chart showing each category of investors as a percentage of total traded value in 2011. I would take roughly half of the proprietary figure and add it to foreigners to undertand the full impact on the market. I find it surprising that the share of Individuals was so high at about 17% in 2011, three times larger than the next largest category of Trust banks. We hear a lot about ‘Mrs. Watanabe’ and how this mythical average Japanese housewife (for in this country the wives are in charge of household purse strings) could reinvigorate the markets if only she could be enticed to invest her considerable savings in stocks. Well, there’s the numerical evidence of Mrs. Watanabe’s importance.

Who’s driven the Japanese market in 2012?

Let’s just remind ourselves of the way the market has performed so far in 2012 by looking at the following daily chart of the TOPIX index. The index peaked at 872.42 on 27 March 2012 and hit a low of 695.51 on 4 June 2012.

TOPIX index year-to-date in 2012

If we take the net sales and purchases for each major category of investors and create a running total for January to August 2012 (we do not yet have monthly data for September), we end up with a figure like the one below. Basically foreigners have been net buyers along with, rather surprisingly, ‘Business companies’, which covers a whole range of normal corporations. As of the end of August 2012, foreigners were cumulative net buyers to the tune of 645.5 billion yen, while at the other end of the scale, trust banks were cumulative net sellers of 393.6 billion yen.

Tokyo Stock Exchange cumulative net buying and selling by investor type in 2012

The graphic above is a snapshot of the state of play at the end of August and doesn’t tell us anything about the evolution of buying and selling through the year in 2012. If we look at the same running total for each month in 2012, from January to August, we can see that there was a dramatic switch in April 2012.

Tokyo Stock Exchange cumulative net traded value January to August 2012 by investor type

Market participants swap behaviour in April

Up until April foreigners were huge buyers, as were ‘proprietary’ participants. Once the market peaked, foreigners and proprietary traders became big sellers, so much so in the case of proprietary that the category become net sellers for the entire year in May 2012. This action is easier to appreciate if we split net traded value into two periods, from January to March and from April to August 2012, as shown in the two charts below. Foreigners bought a cumulative net 1,289.8 billion yen up to the end of March, then were cumulative net sellers to the tune of 644.3 billion from April to August. This was not the biggest net seller category: that dubious honour went to ‘proprietary’ at -1,059.9 billion yen from April to August (although foreigners are probably very well represented in this category also).

Tokyo Stock Exchange cumulative net traded value January to March 2012

Meanwhile, trust banks – cumulative net sellers of 841.7 billion in January to March – bought a cumulative net amount of 448.1 billion yen from April to August 2012. Individuals sold the market all the way up (-783.2 billion) in January to March then bought it all the way down in the April to August period, becoming cumulative buyers of 391.8 billion during that time.

So, it sucks to be an individual investor. This comment is more than flippancy on my part: Japanese individual investors have the power to move the market upward, but consistently losing money is not going to motivate them.

Tokyo Stock Exchange cumulative net traded value April to August 2012


You may wish to know more about the definitions for each category. I can do no better than to quote the TSE’s own web page here.

(1) Foreigners
a. “Non-residents” as defined in Article 6, Paragraph 1, Item 6 of the Foreign Exchange Act (Foreign Exchange and Foreign Trade Act). Since the overseas branch offices and overseas subsidiaries of Japanese corporations are also classified as “Non-residents”, they are included in “Foreigners”, but since Japanese branch offices of foreign corporations excluding those in b. below are classified as “Residents”, they will be included in (5) Other corporations or (9) Other financial institutions. Similarly, since Japanese subsidiaries of foreign corporations are classified as “Residents”, they will be classified into the respective investment category.
b. Japanese branch offices of foreign securities companies which are not trading participants on TSE.

(2) Securities companies
Brokerage orders from other securities companies excluding 1(b). Proprietary trading by all trading participants covered by the survey is aggregated under “Proprietary”, not “Securities cos.”
(3) Investment trusts
Investment trust management companies and asset management companies as defined by the Investment Trust Act (Act on Investment Trusts and Investment Corporations).
(4) Business companies
Joint-stock companies, limited companies and partnership companies (mochibun kaisha), including general partnership companies (gomei kaisha), limited partnership companies (goshi kaisha) and limited liability companies (godo kaisha), that do not fall under (1) through (3) inclusive and (6) through (9) inclusive. Also includes business companies that control financial institutions. All holding companies fall under business companies.
(5) Other corporations
Institutions excluding (1) that do not fall under financial institutions, investment trusts and business companies. More specifically, groups such as government and local public bodies and their affiliated organizations, foundations, special corporations, employee shareholding associations, friendship associations and labor unions, and Japanese branch offices of foreign corporations that are not financial institutions fall under this category.
(6) Life & non-life
Life insurance companies and non-life insurance companies as defined by the Insurance Business Act.
(7) City & regional banks
Regular Japanese banks licensed under the Banking Act.
(8) Trust banks
Trust banks that are members of the Trust Companies Association of Japan.
(9) Others financial institutions
Financial institutions besides (6) through (8) inclusive.
More specifically, credit unions, credit associations, agriculture-related financial institutions, various mutual aid associations, government-related financial institutions, and Japanese branch offices of foreign corporations fall under this category.

This is all pretty straightforward. You may have raised an eyebrow at the thought of Japanese branch offices of foreign corporations coming under the ‘Others financial institutions’ category, but it turns out that this category accounts for such a small part of total trading that this definitional quirk cannot be an issue. I will take another look, but I suspect that Japanese branch offices are usually not the entity reported in the trade these days and that most orders from foreigners for Japanese stocks are routed through Singapore or Hong Kong.

Previous post:

Next post: