Costly index investing in foreign markets

Journal of Financial Markets xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Financial Markets journal homepage: www.elsevier.com/locate/finmar

Costly index investing in foreign markets☆ Alvaro Pedraza a, ∗ , Fredy Pulga b , Jose Vasquez c a

World Bank Group, 1818 H Street NW, Washington D.C. 20433, USA Universidad de la Sabana, Km 7 Autopista Norte, Bogota, Colombia c World Bank and School of Economics, Navarra University, Garcia de Paredes 55, 28010, Madrid, Spain b

article info

abstract

Article history: Received 4 February 2019 Revised 12 September 2019 Accepted 17 September 2019 Available online XXX

We study trading behavior and performance of foreign investors by level of active management. Using a comprehensive Colombian dataset with complete transaction records, we find that aggregate underperformance of foreign investors is attributable to passively-managed foreign funds. These funds pay higher prices to increase the speed of their trades in order to accommodate daily flows proportionally to their benchmark index. Higher transaction costs occur on days when they trade multiple stocks in the same direction and make large trades near market closing. The findings highlight the potential costs of index investing in developing countries or in securities with low trading activity. © 2019 Published by Elsevier B.V.

JEL classification: F36 G11 G15 G23 Keywords: Foreign investors Index funds Performance Transactional data

1. Introduction “Passive investing has become investors default, driving billions into funds that track indexes. It’s transforming Wall Street, corporate boardrooms and the life of the neighborhood broker.” (Anne Tergesen and Jason Zweig, The Wall Street Journal, October 17, 2016) The relative performance of foreign versus domestic investors has been of long-standing interest to financial economists. Whether foreign investors and foreign money managers under-perform local investors is often attributed to an information advantage of one group over the other. For instance, domestic investors might have an edge due to cultural affinity and famil-

☆ We are grateful to Paolo Pasquariello and two anonymous referees whose comments substantially improved the paper. We have also benefited from the comments of Sushant Acharya, Pablo Cuba, Robert Cull, Filippo De Marco, Aart Kraay, Pete Kyle, Soledad Martinez, Claudia Ruiz, Giorgo Sertsios and Rafael Zambrana, as well as the seminar participants at Texas A&M University, EAFIT University, the 25th Finance Forum, 2017 AFFI International Conference, and the 22nd EBES Conference. We thank Adriana Cardenas and other members of the Colombian Stock Exchange Research Committee for providing the data and for multiple discussions. We acknowledge financial support from the Strategic Research Program. A previous version of this paper was circulated under the title “Do foreign investors underperform? An empirical decomposition into style and flows.” The views expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.

∗ Corresponding author. E-mail addresses: [email protected] (A. Pedraza), [email protected] (F. Pulga), [email protected] (J. Vasquez).

https://doi.org/10.1016/j.finmar.2019.100509 1386-4181/© 2019 Published by Elsevier B.V.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

2

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

iarity with local conditions (e.g., Chan et al., 2005; Portes and Helene Rey, 2005; Leuz et al., 2010). In fact, the international finance literature has often used the view that distance reduces the quality of information to explain home bias, the volatility of capital flows, and herding (Brennan and Cao, 1997; Lewis, 1999; Karolyi and René, 2003). Conversely, foreign investors might have an advantage if they are more experienced or if they acquire private information from their presence in multiple markets (Froot and Ramadorai, 2008; Albuquerque et al., 2009). Average performance, however, does not only depend on skills in identifying investment opportunities, but also on various constraints faced by fund managers such as the manager’s discretion to accommodate flows if the fund is actively or passively managed (index funds). In this paper, we investigate the relation between the level of active management (active vs. passive funds), distance (foreign vs. domestic investors), and trading performance using a unique data set of transactions from the Colombian Stock Exchange. The advantage and novel aspect of the data is that for all the transactions in the market between January 2, 2006 and January 29, 2016, we have a unique investor identifier that allows us to track each fund or individual over time. Given the panel structure of the data, we are not only able to compare average performance between foreign and domestic investors, but we can also study the extent to which trading performance is related to investor/fund characteristics, such as the level of active management. Despite the vast segment of the literature that focuses on investors who are separated by borders, to the best of our knowledge, we are the first to use information on active versus passive strategies, together with complete transaction histories, to analyze the relative skill and performance of different types of foreign versus local investors. We find that, on average, foreign investors trade at unfavorable prices relative to domestic investors, paying higher prices for stock buys and receiving less when they sell. While our baseline estimates are in line with previous findings in other countries (e.g., Choe et al., 2005, for Korea),1 our access to investor-level trading records allows us to depart from previous work and examine the question of relative trading skills and performance from a new angle, namely, by level of active management. In particular, our cross-section analysis reveals that the disadvantage of foreign investors is largely attributable to passivelymanaged foreign funds, that is, those whose strategy consist of replicating the return of an index by buying and holding all (or almost all) index stocks in the official index proportions. These include explicit index funds and closet indexers (a term commonly used for self-declared active funds that are largely passively managed). Furthermore, we find that the worst trades are on days when these funds (i) trade multiple stocks in the same direction, (ii) buy (sell) the same stock multiple times, and (iii) make large trades at the end of the trading session. Taken together, our evidence indicates that the inferior performance of passively managed foreign funds tends to be on days when managers trade more intensively to accommodate flows in similar proportions to their benchmark index. To further analyze the role of active management on trading performance, we compare passively-managed foreign funds versus domestic funds with similar strategies. We find that relative to passively-managed domestic funds, passively-managed foreign funds buy and sell stocks at unfavorable prices even after controlling for several measures of daily trading intensity. This result suggests that (passive) index investing is not a sufficient condition to explain the under-performance of foreign investors. We explore three non-mutually exclusive explanations for our findings. First, if managers of passively-managed foreign funds are better informed, their trades have a larger permanent price impact. Second, with incomplete risk-sharing, large trades would temporarily disturb the position of liquidity providers, and investors pay compensation in the form of higher prices — the so-called “inventory effects.” Third, passively-managed foreign funds might choose to trade after prices have moved if they follow intra-day momentum strategies. To distinguish between information, liquidity, and momentum explanations, we examine the joint dynamics of returns and net equity flows by level of active management. We find that large purchases by passively-managed funds lead to same-day price increases, followed by a significant price decline the following day. The evidence suggests that these funds pay higher transaction costs (e.g., their trades have a larger temporary price impact) and does not support the hypothesis that they are better informed than domestic institutions, or that these trading costs result from intra-day momentum. Finally, since global and regional investment strategies between local and foreign investors might be different, we examine the response of investors to international markets returns. We find that passively-managed foreign funds trade more intensively in Colombian equity on the same day of large price changes in international markets. In particular, net equity flows towards Colombian stocks increase with high stock returns in Latin America and other emerging markets, and decrease with large returns on developed markets. On the contrary, active foreign funds do not have a systematic response to asset price shocks outside Colombia. Overall, the evidence seems to suggest that passively-managed foreign funds have a strong contemporaneous response to global and regional returns, and in turn, experience large transaction costs when trying to accommodate daily flows. While the effect of distance on performance has been extensively studied in the literature, the role that active management plays on the investment strategies of foreign asset managers has received little attention. A growing number of empirical studies use complete transaction records within a country and compare foreign-managed versus domestic-managed funds. Some of these studies provide support in favor of the superior skill of domestic investors (Hau, 2001; Choe et al., 2005; Dvoˇrák and AuthorAnonymous, 2005; Agudelo et al., 2019), while others document that foreign investors outperform locals (Grinblatt and Keloharju, 2000; Barber et al., 2009), and some find no differences in performance (Seasholes and Zhu, 2010). Most transactional data sets, however, do not identify investors separately. Instead, stock buys and sells are aggregated by investor type (foreign or

1

Agarwal et al. (2009) also has similar aggregate results for Indonesia over an eight-year period.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

3

domestic).2 Given these limitations in the data, many studies do not account for within group heterogeneity that might explain trading gains or losses, and what they often estimate is an average performance gap between local and foreign investors. Given the unique investor ID in our transactional database, we can overcome previous data restrictions and study relative trading performance at a more granular level. More precisely, we add to the literature by examining the performance of foreign investors by level of active management. In our sample, foreign funds with passive strategies looking to accommodate daily flows have a high demand for immediacy.3 Consequently, the managers of these funds end up paying higher transaction costs and display inferior returns. In contrast, foreign funds with more discretion over how they accommodate daily flows trade at fair prices — buying and selling stocks at average daily prices. Moreover, actively-managed foreign funds display similar returns than domestic funds with active strategies. Collectively, our cross-sectional analysis provides a detailed portrait of the performance of foreign investors, with passively-managed funds largely underperforming other investors. Our analysis also contributes to three other strands of the literature. First, our work is related to the empirical literature on the behavior of international equity flows and their connection to local asset returns. Prior research often used high-frequency data on aggregated flows to examine positive feedback trading, momentum trading, or contagion trading among foreign investors (Choe et al., 1999; Froot et al., 2001; Griffin et al., 2004; Kaminsky et al., 2004). Since we track each investor (individual or fund) over time, we are able study the relation between stock prices and foreign equity flows for investors with passive and active strategies. We find that flows by passive funds generate a large temporary price impact. In all, our evidence shows that the distribution of management styles across foreign investors can provide valuable insight for a policymaker or a monetary authority concerned about the potentially destabilizing effects of foreign portfolio flows. Although our paper presents new findings using rich data from a single country, other global studies suggest that funds worldwide are becoming less active and indexing strategies are more common among investors (Cremers et al., 2016; Williams et al., 2017). Therefore, the types of behavior documented here and their implications for portfolio performance and asset prices are likely to be found in other countries. Second, our evidence is related to the literature on the relative value of active versus passive management. A large number of papers document a poor track record for active funds, with average returns for investors significantly below those of passive benchmarks (e.g., Wermers, 2000; Bollen and Busse, 2001; Pástor and Stambaugh, 2002; Avramov, 2006). More recently, Cremers and Petajisto (2009) find that managers with more active trading strategies exhibit relatively more skill than less active managers. We contribute to the literature by examining intra-day trading patterns by level of active management and by documenting a disadvantage among passive foreign investors — the cost of index investing in a small equity market. While it is well known that investment styles significantly influence institutional trading costs (Chan and Lakonishok, 1995; Keim and Madhavan, 1997; Yan and Liquidity, 2008) and that bad timing of flows can erode a fund’s performance (Edelen and Roger, 1999; Greene and Hodges, 2002), we provide the first evidence that mechanical passive investment strategies produces significant price-pressure and results in worse trading performance. We would expect similar findings in other developing countries, or in small (illiquid) stocks in developed countries. Finally, our paper is related to the literature on the asset price effects from passive investing. For example, when a stock is added to a popular index, it exhibits abnormal returns (Chen et al., 2004a; Greenwood, 2005; Hau, 2011) and experiences increased return co-movement with the rest of the stocks in the index (Barberis et al., 2005; Greenwood, 2008). Such effects are assumed to derive from the shifts in the demands of investors that benchmark their strategies against the index, although their actual trading behavior is not directly observed. In this paper, given our comprehensive data, we test for price effects by examining the joint dynamics of stock returns and net purchases by funds with different levels of active management. In our setting, passively-managed foreign funds incur large transaction costs in the form of a large temporary price impact. They appear to pay liquidity providers to increase the speed of their trades to stay close to their benchmark before the end of each trading session. The rest of the paper is organized as follows. In Section 2, we provide some background on the Colombian institutional setting and describe the data. In Section 3, we document the general trading patterns across different investor types. We study the relation between distance, management style, and performance in Section 4 and analyze the role of flows on performance in Section 5. We conclude in Section 6. 2. Background and data 2.1. The Colombian Stock Exchange In contrast to fragmented stock markets in the United States, the Colombian Stock Exchange (CSE) operates as a single electronic central limit order book for each stock registered in the Exchange. Our data are from January 2, 2006 to January 29, 2016. During this period, the CSE implemented several changes in order to meet international standards. Before February 2009,

2 This is typically a consequence of data restrictions in order to guarantee the anonymity of each trader. The exceptions are Hau (2001) and Seasholes and Zhu (2010). Hau (2001) uses data on a group of professional traders but does not study the role of active versus passive management. Seasholes and Zhu (2010), on the other hand, focus on individual investors. 3 Barber et al. (2009) and Agarwal et al. (2009) find that differences in performance across domestic and foreign investors can be traced to the aggressive orders of each group, e.g., buy limit orders with high prices and sell limit orders with low prices. These papers, however, do not distinguish investors by level of active management.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

4

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

the market operated as a continuous trading system from 8:00 to 13:00 (local time) on weekdays where participants were only allowed to place limit orders. After February 9, 2009, a new trading platform was launched, allowing market and stop orders. In addition, a batch auction was introduced during the last 5 min of each trading day. During this interval, trades are collected and the system estimates an adjudication price at the maximum volume allocated among the tenders, with the auction closing randomly in ± 60 s around the daily closing time. The auction price is also the daily closing price for each stock. Starting January 3, 2011, the trading session was extended from 8:30 to 15:00 with the other rules remaining the same. Finally, on February 6, 2012, the market synchronized its daily trading session with the New York Stock Exchange, from 9:30 a.m. EST to 4:00 p.m. EST.4 Trades are submitted via authorized brokers registered at the exchange and buy and sell orders meet via the Automated Trading System. Short-sells were allowed during our entire sample, however, investors were required to close any short position by the end of each trading day.5 There are no individual or aggregate ownership limits on foreign investors, and they are allowed to either reinvest or transfer earnings such as dividends or capital gains with little restrictions. Since the central bank has discretionary control on net currency flows, including those to equity instruments, foreign investors are required to report to the monetary authority their initial investments and any subsequent transactions that involve any currency exchange. 2.2. Data The database, which includes all the transactions in Colombian stocks, was provided by the CSE. The data include the date and time of each transaction, a stock identifier, order type (buy or sell), transaction price, number of shares, and broker ID. The key novel aspect of the data, and most importantly for our analysis, is that every transaction record has a unique investor ID number that allows us to track all the transactions for each investor throughout the entire sample period. Moreover, the CSE provided end of the year holdings in Colombian stocks for each investor. The data also includes whether an investor is foreign or domestic, and whether it is an index fund or an exchange-traded fund (ETF).6 Domestic investors are further classified into individuals or institutions. Institutions can be broadly categorized into corporations, pension funds, mutual funds, banks, insurance companies, and brokerage firms trading on their own accounts. It is important to point out that our database does not allow us to distinguish whether a non-index foreign investor is an individual or an institutional investor. Based on this, for the rest of the paper, we sometimes refer to foreign investors interchangeably as funds.7 There were over 13 million transactions during our 10-year sample period. Panel A in Table 1 presents the share of trading volume attributable to different investor types and the number of investors over two sub-samples: 2006–2010 and 2011–2016. Domestic institutions dominate the CSE, accounting for just over half of the total value traded in both periods. Corporations, pension funds, mutual funds, and brokerage firms account for 98% of this volume over the entire sample, with banks, insurance companies, and other institutions representing a small fraction of the trading activity of domestic institutions. Consistent with international trends, there is a growing presence of foreign investors over time. The number of foreign investors nearly doubled from 1440 to 2,830,8 and the share of total traded value by these group increased from 3.8% in the first half of the sample period to 19.1% in the second half. Fig. 1 shows the time series of the share of traded value by investor type. The figure highlights the increasing importance of foreign investors in recent years. Panel B in Table 1 shows that the average number of trades by domestic institutions and foreign investors are 317 and 316 respectively. The distribution of trades among different institutions, however, is quite heterogeneous, with brokerage firms and mutual funds conducting the largest number of trades on average. We restrict our analysis by considering only those investors who traded at least a 100 times over the 10-year period.9 We collect stock and market data from Bloomberg, Datastream and the Superintendencia Financiera de Colombia (Colombian Financial Superintendency, SFC). We obtain stocks returns, share price, trading volume, and bid-ask spread from Bloomberg and book value of equity and shares outstanding from Datastream and the SFC. The number of traded stocks fluctuated between 62 and 71 during our entire sample period.10 In Table 2, for every year in the sample, we report the market rate of return, end-of-year market capitalization, total flows by foreign investors, and number of stocks. According to the table, the performance of the CSE closely mimicked that of other stock markets in Latin America. The table is also indicative of increased participation by foreign investors in the CSE. Despite the

4

For non-liquid stocks, an opening auction starts 15 min earlier, but trading at this time represents less than 0.01% of the daily traded value. Other rules include a suspension mechanism on each stock based on volatility relative to a “base price” published on a daily basis. When a stock price changes more than 10% relative to its reference price, the exchange suspends trading in that stock temporarily for 30 min. In addition, when the representative stock index (Colombian Capitalization Index) decreases more than 10% during a trading session, all transactions are suspended until the next trading day. 6 The CSE does not track explicitly index funds, nor does it have funds’ prospectus. Their records, however, do contain names of each investor — these were not shared with us to maintain annonymity. To identify index funds, we provided the CSE with a list of international ETFs and self-declared index funds from Factset Equity Ownership database, and a list of domestic mutual funds that according to the financial supervisory agency, SFC, were index funds. The CSE then classified funds in the transactional database as index or ETFs if there was a perfect match with our list, or if the word “index” or “ETF” appeared in any part of the fund’s name. 7 Informal conversations with registered brokers suggest that most (and in several cases all) of their foreign clients are institutions. 8 The corresponding numbers for domestic institutions were 6834 and 10,344 respectively. 9 We are aware that this cut might bias the sample to those investors who have been active in the market for a longer period. We performed our analysis by alternatively reducing the threshold to 50 trades and also for a higher threshold of 200 trades. These changes do not affect any our findings. 10 In calculating these numbers, we treat ordinary and preferred shares issued by the same firm as different stocks. 5

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

5

Table 1 Traded value and number of investors. Panel A presents the share of total traded value and number of investors by each investor type in the database. Share of traded value is calculated relative to the total amount traded in the period. Number of investors that made at least one transaction during the period. Panel B displays distribution statistics of the number of trades by investor type.

Panel A. Traded value and number of investors Investor Type

% of Total Value

Number of Investors

2006–2010

2011–2016

2006–2010

2011–2016

46 50 16 13 6 14 2 4

27 54 13 12 6 21 2 19

265,395 6823 6270 32 268 51 202 1440

340,644 10,339 9327 50 421 50 491 2830

Domestic Individuals Domestic Institutions Corporations Pension Funds Mutual Funds Brokerage Firms Others Foreign

Panel B. Distribution of number of trades by investor type Investor Type

Average

p5

p25

p50

p75

p95

p99

Max

Domestic Individuals Domestic Institutions Corporations Pension Funds Mutual Funds Brokerage Firms Others Foreign

14 317 110 8320 1048 24,965 161 316

1 1 1 8 1 2 1 1

1 2 2 313 3 32 1 2

1 8 7 1839 15 345 5 11

3 42 39 8986 216 7936 31 81

37 536 405 44,119 4233 135,067 383 1090

187 2064 1843 51,056 15,256 421,236 2786 4922

57,913 421,236 39,976 51,056 62,654 421,236 23,209 92,774

Fig. 1. Trading by investor group. Percentage of monthly traded value by investor type relative to the total traded value of the month.

large amount of outflows by foreign investors during 2014 and 2015, accumulated foreign net flows are positive over the entire sample period. 3. Investor/fund style 3.1. Active versus passive management There are 151 foreign index funds and 180 domestic index funds in our sample. In addition to identifying these self-declared index funds, we classify all investors in our sample by their level of active management in Colombian stocks. To be precise, we Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

6

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx Table 2 Descriptive statistics for the Colombian Stock Exchange. The table displays yearly market stock returns. For Colombia, returns are calculated using the COLCAP, a value-weighted index. The MSCI Latin America index captures large cap and mid cap firms across five countries: Brazil, Chile, Colombia, Mexico, and Peru. On December 31, 2015, the country weight of Colombia in the index was 3.15%. Market capitalization (market cap) is the end-of-year total stock market value. Foreign flows and number of stocks are from the CSE dataset.

Year

MSCI Latin America

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

43.2% 50.4% −51.4% 103.8% 14.7% −19.4% 8.7% −13.4% −12.3% −31.0%

Colombian Stock Market

Market Cap ($ billion cop)

Foreign Flows ($ billion cop)

Number of stocks

16.0%

11,161 10,694 7561 11,602 15,497 12,666 14,716 13,071 11,635 8547

642 275 631 943 122 −13 −44 479 −370 −1390

68 67 66 62 63 71 72 70 66 65

−4.3% −34.7% 42.8% 28.9% −20.2% 15.0% −11.9% −11.6% −30.8%

compute Active Share (Cremers and Petajisto, 2009) for each investor or fund in our sample as: Active Sharei,t =

N 1∑| | |w − windex,s,t | , | 2 s=1 | i,s,t

(1)

where wi,s,t and windex,s,t are the portfolio weights of stock s in fund i and the benchmark index at time t, respectively, and the sum is taken over the universe of Colombian equity holdings. We compute this measure by investor/fund at monthly frequency. For the benchmark index, we use the two most popular Colombian equity indices that track the overall market performance COLCAP, which is a value-weighted index, and IGBC, which is a liquidity and value-weighted index. We compute equation (1) using each of these indices separately and then set the investor’s Active Share to be the minimum of the two values.11 There is an important limitation from our measurement. Here, Active Share only captures the level of active management in Colombian stocks. For instance, an international fund in our sample might look very passive in its equity investments in Colombia, and at the same time hold an overall active strategy, with total stock holdings significantly different from widely followed international equity benchmarks.12 In turn, when we measure Active Share for each fund i in the sample, the level of active management refers exclusively to the fund’s strategy within Colombian stocks. We classify investors/funds into three groups according to their management strategies: (1) index funds, as identified by the CSE. (2) Passive, referring to those investors or funds that are not identified as index trackers, but with holdings in Colombian equity similar to an index. More precisely, these are non-index funds with Active Share below 60%.13 (3) Active, that is, nonindex funds with Active Share above 60% — or Colombian stock holdings that are significantly different than common indices. Table 3 reports the share of assets under management (AUM) in Colombian stocks for the three groups of active management strategies. Consistent with global trends, the share of total assets managed by passive investors increased in the second half of the sample. Among foreign investors, the share of index funds rose from 4.8% to 21.8%, and that of non-index funds with passive strategies increased from 22.9% to 42.8%. Within domestic institutions, only some mutual funds are clasified as explicit index funds (these include domestic ETFs). However, other domestic asset managers, such as pension funds, largely fall under the category of passively managed.14 Given the unique nature of our transaction database with investor ID, we document how passive and active investors trade during the day. In Table 4, we summarize the intra-day profile of traded value when we divide the day into eight time intervals.15 Panel A in the table reports the proportion of the traded value by each group relative to the total traded value in the interval. Excluding the closing batch, domestic institutions account for more than half of the value traded during each period, with foreign investors increasing their relative importance in the afternoons. In the closing batch, foreign investors represent 40.3% of the interval, with index and passive funds accounting for most of those trades, 12.1% and 19.5% respectively. Panel B in the

11 Using Active Share to characterize management style has an advantage over using self-declared fund style, which may not accurately reflect actual investment behavior. For example, Cremers et al. (2016) find that 20% of worldwide mutual fund assets are managed by closet indexers — investors who declared themselves as active, but in reality, tracked their respective benchmarks closely as passive investors would have. 12 Alternatively, a fund could hold Colombian stocks in different proportions than the domestic index, yielding a high Active Share measure, while at the same time following a passive strategy in the rest of its portfolio. 13 This threshold has been commonly used in the literature to classify portfolio management as active or passive (Cremers et al., 2016). All the results in our paper are robust when we reduce the threshold to 50%. Our main empirical findings also hold when we split our groups of active management strategies by quartiles of Active Share. 14 According to Pedraza (2015), Colombian pension funds face a financial penalty if they fail to generate a minimum return relative to a peer- and market-based benchmark. Hence, it is not surprising that the portfolios of these funds replicate to a large extent the market index. 15 The sample includes the period between February 6, 2012, and January 29, 2016 when the CSE synchronized its operations with the NYSE. The table is quantitatively similar if we include the period from February 9, 2011, to February 6, 2012, when the market traded from 8:30 to 15:00 h and if we match the trading session.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

7

Table 3 Passive investors. This table reports the share of assets under management for index and passive funds as a proportion of the total assets for each group. The sample is divided in two five-year periods: 2006–2010 and 2011–2016. Passive investors/funds are defined as non-index funds with an Active Share below 60%. Sample includes investors with at least 100 trades.

Investor Type

Domestic Institutions Corporations Pension Funds Mutual Funds Brokerage Firms Others Foreign

2006–2010

2011–2016

Index

Passive

Index

Passive

4.3 – – 20.5 – – 4.8

49.5 13.6 97.8 70.1 45.6 15.8 22.9

8.3 – – 35.7 – – 21.8

63.7 15.2 85.4 50.2 74.4 22.5 42.8

Table 4 Average proportion (%) of intra-day trading volume for each type of investor. This table reports the average percentage of intra-day traded values by level of active management: index, passive, and active funds. Passively-managed (actively-managed) funds are non-index funds with Active Share below (above) 60%. The percentages are relative to the total trading value for the interval (Panel A) and relative to the total traded value of the day by each group (Panel B). The sample in Panels A and B include the period from February 9, 2012 to January 30, 2016, when the Colombian Stock Exchange synchronized its trading day with the New York Stock Exchange. Panel C presents the traded value of foreign index and passive funds in the closing auction.

Trading intervals

Individuals

Domestic Institutions Index

Foreign

Passive

Active

Index

Passive

Active

24.8 26.8 28.1 28.0 26.0 23.7 24.1 16.2

19.5 19.1 18.8 18.2 18.1 17.6 17.4 11.0

1.8 2.1 2.3 2.6 3.5 4.3 3.9 12.1

3.8 4.9 5.6 6.1 6.8 7.8 7.0 19.5

8.1 9.5 9.7 9.9 11.9 11.5 10.3 8.7

Panel B. Relative to the traded value of the day by investor group 9:30–10:30 22.4 16.8 18.0 10:30–11:30 17.2 17.9 18.7 11:30–12:30 14.5 15.6 18.2 12:30–13:30 11.3 12.3 14.0 13:30–14:30 7.2 6.8 7.9 14:30–15:30 9.1 8.9 8.0 15:30–15:55 7.4 7.6 7.0 Closing batch 10.9 14.1 8.2

20.8 18.3 16.2 12.5 7.5 8.7 7.0 9.0

9.9 11.7 10.0 7.8 6.8 8.6 5.8 39.3

10.0 10.3 9.8 9.3 5.8 8.5 6.9 39.5

15.1 18.1 15.6 11.5 8.9 10.1 7.2 13.6

Panel C. Foreign investors in the closing auction 2009 2010

2011

2012

2013

2014

2015

Index Relative to interval Relative to day

1.3 39.3

5.2 44.0

6.3 41.9

10.8 34.8

13.0 32.4

14.5 46.0

12.6 49.1

Passive Relative to interval Relative to day

5.7 24.5

7.2 27.5

15.0 34.2

18.9 34.5

24.6 48.2

29.3 37.0

29.9 40.2

Panel A. Relative to the trade value of the interval 9:30–10:30 38.5 3.5 10:30–11:30 33.5 4.0 11:30–12:30 31.2 4.3 12:30–13:30 30.9 4.4 13:30–14:30 29.6 4.0 14:30–15:30 30.6 4.5 15:30–15:55 32.4 4.9 Closing batch 26.7 5.7

table reports the proportion of traded value in each interval relative to the total daily value traded by investor group. Domestic institutions and individuals execute most of their trades in the early hours of the day. In contrast, index and passively-manged foreign funds trade over 39% of their daily value in the closing batch. While this behavior is more pronounced for the latter years in the sample, passively-managed foreign funds were consistently very active in the last 5 min of the trading session since the closing batch auction was introduced in 2009 (Table 4, Panel C). Trading at or near closing prices might be an efficient strategy for open-end funds. For example, mutual funds are “forward priced,” which means that while investors can place orders to buy or sell shares throughout the day, these orders will be executed at the same price (i.e. the end-of-day net asset value of the fund). A fund manager that trades at closing prices mitigates the timing risk due to non-simultaneous purchases or sales of the fund shares and the underlying portfolio. In this scenario, the price that new shareholders pay for shares that are created coincides with the value of the underlying securities of the fund.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

8

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx Table 5 Fund characteristics by level of Active Share. This table reports mean values for the variables across foreign and domestic investors, grouped by type of active management: index funds, passive (non-index funds with Active Share below 60%), and active funds (non-index funds with Active Share above 60%). Number of stocks denotes the different number of stocks held by each investor/fund by month. Stocks traded are the different number of stocks traded by each investor/fund during a month. Monthly net flows are normalized by the value of the investor’s portfolio in the previous month. Percentage of investors in groups is the proportion of each investor type in the interval relative to the total number of investors in each type.

Index

Passive

Active

Foreign Active Share Assets ($ billion cop) Fund turnover ratio Number of stocks Stocks traded Net flows % of investors in the group % of total number of trades

48.2 56.3 0.3 12.5 6.0 0.5 7.5 17.2

46.3 31.2 0.4 14.2 5.0 0.2 26.9 38.7

82.6 10.8 0.9 6.5 2.0 0.0 65.6 44.1

Domestic Institutions Active Share Assets ($ billion cop) Fund turnover ratio Number of stocks Stocks traded Net flows % of investors in the group % of total number of trades

43.9 21.7 1.5 20.5 7.3 0.0 3.1 9.6

56.9 48.4 0.9 16.1 3.7 −0.1 26.4 58.5

79.3 9.3 1.5 9.5 2.1 −0.2 70.6 31.8

Conversely, the price that old shareholders receive for their redemptions matches the value of the underlying securities in the fund. Interestingly, only passively-manged foreign funds display such intensive trading at the end of the trading session. For example, while foreign index funds trade 39.3% of their daily traded value in the closing batch, domestic index funds only trade 14.1% of their volume during this time. It is possible that the dollar value of Colombian stock holdings for foreign index funds and passively-mananged foreign funds is small relative to their total AUM. For instance, the average weight of Colombia in the MSCI Latin America index was 3% during our sample period. If the share of the portfolio of these passively-managed foreign funds is smaller than other foreign investors, it is likely that the costs associated with engaging in strategic trading throughout the day (e.g., the cost of monitoring intra-day market conditions) might outweigh the benefits. In the rest of the paper, we examine the implication of these trading patterns on performance.

3.2. Measures of trading activity In addition to measuring the investor/fund management style by its self-declared strategy and Active Share, we employ other common proxies of trading activity. We measure the fund turnover ratio as the minimum of aggregate sales and aggregate purchases of stocks, divided by the average of the net asset value of the fund. We calculate the fund turnover ratio at yearly frequencies for each investor/fund in the database. We also use equity flows normalized by the total AUM at the beginning of each month. For each investor/fund, we report the different number of stocks held in the portfolio and the different number of stocks traded per period. Table 5 reports the means of the relevant variables across foreign and domestic investors, for each group of active management — index, passive, and active. According to the table, the average foreign index fund manages a portfolio of 56.3 billion pesos (17.7 million USD) with yearly turnover of 0.3, has monthly net flows of 0.5% relative to the AUM (6% by year), holds 12.5 different stocks, and trades 6 different stocks every month. While these funds only represent 7.5% of the number of foreign investors in Colombia, their number of trades account for over 17% of the total trades by foreign investors. Together with passive funds, these two groups represent 53.9% of all transactions among foreign investors. On the other hand, active funds are smaller, display higher turnover, and hold and trade less stocks on average. 4. Investor performance In order to test whether foreign investors are at a disadvantage, we measure trading performance as the value-weightedaverage-price (VWAP). The VWAP for each stock s, on day d by investor i is measured as: VWAPi,s,d =

Bi,s,d As,d

× 100,

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

(2)

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

9

where Bi,s,d is the volume-weighted average price for all stock buys and sells separately for each investor i, and As,d is the volumeweighted average price for all trades. This price ratio is computed for buys and sells separately and captures how much more a particular investor paid (received) relative to the average price of the day when she buys (sells). The VWAP measure offers at least two advantages. First, it captures short-term market timing ability and does not require the definition of an asset pricing model. Second, since the measure has been used by Choe et al. (2005) and Agarwal et al. (2009), we can directly compare our results to those in these studies and extend previous findings by decomposing performance into management styles while controlling for other investor/fund characteristics.16 One limitation to the VWAP, however, is that by comparing trading among investors at daily frequency, the measure fails to capture performance at longer horizons. More precisely, the measure does not identify whether a purchase that was initially executed at a price above the average daily price turns out to be profitable during the investor’s holding period. Moreover, if a trade was part of a strategy by the investor that involves several stocks, the individual trade might be deemed as inferior but the overall strategy could be successful. In order to consider total portfolio performance, we also measure risk-adjusted returns by investor/fund style in Subsection 4.2. Averages of VWAP by investor type are presented Panel A of Table 6. Foreign investors pay 4 bps more than the average daily price when they buy stocks, while domestic institutions and domestic individuals pay 4 and 8 bps less than the average price respectively. The 8 and 12 bps differences between foreign investors and both domestic institutional investors and individual investors are statistically significant. For sells, foreign investors receive 7 bps less than domestic institutional investors and 8 bps relative to domestic individual investors. These results indicate that on a round-trip trade, foreign investors face greater transactions costs on the order of 15 bps compared with domestic institutions and 20 bps with respect to domestic individuals.17 Furthermore, an investor who trades six times per year would contemplate a drag on performance in excess of 1% of her total traded value. Given the yearly average buying value of foreign investors of 4.3 trillion COP (1.4 billion USD) and their yearly selling value of 3.4 trillion COP (1.06 billion USD), their under-performance is equivalent to paying 6.2 billion COP (2.0 million USD) every year on transaction costs. The documented disadvantage is quantitatively similar to that reported by Agarwal et al. (2009) for Indonesia over an eight-year period, and smaller but in the same order of magnitude to the one reported by Choe et al. (2005) for Korea between 1996 and 1998. Panel B of Table 6 reports the average price of transactions by foreign investors sorted by type of active management strategy. The panel presents one of the main findings of our paper. It shows that the inferior trading performance of foreign investors in our sample is attributed to funds that are passively-managed, that is, index funds and those with holdings similar to the market index. In particular, foreign index funds pay the highest price for stock buys while receiving the lowest price for their sells, 11 and 9 bps respectively. On the contrary, actively-managed foreign funds do not trade at prices significantly different from the average price of the day, with daily buys and sells insignificantly different from 100. Given the distinct patterns of intra-day trading activity between passive and active foreign investors documented in Table 4, we report average transaction prices by trading interval. Panel C in Table 6 reports the average trading price of transactions executed in the closing batch and those that are executed during the continuous trading session.18 According to the table, index and passively-managed foreign funds buy (sell) stocks above (below) the daily average price in both intervals. However, these investors trade at significantly worse prices during the closing batch. For example, index funds pay 16 bps over the average daily price for buys during the closing batch. In other words, index and passively-managed foreign funds pay more precisely in the interval when they are trading more intensively. On the contrary, actively-managed foreign funds buy at favorable prices during the continuous trading session (2 bps below the average price of the day) and sell at “fair” prices. These funds, however, do appear to trade at worse prices at the closing, but as reported earlier, total trades during this interval only represent a small fraction of the trades by this group. If foreign index and passively-managed foreign funds are consistently making inferior trades, investors on the other side of these transactions must be receiving high prices for stock sells and paying low prices for stock buys. According to Panel C of Table 6, both passively- and actively-managed domestic funds are making superior trades during the continuous session and at market closing. It appears that domestic institutions, independently from their level of active management, benefit from their trading with foreign funds. For example, while foreign index funds pay 16 bps more for stock buys during the closing auction, sells of index and active domestic funds are 14 and 20 bps respectively above the daily price in the same interval. One potential explanation to our results is that the underperformance of foreign funds could simply reflect the U-shaped trading pattern concentrated at the beginning and end of the day that has been commonly reported in the market microstructure literature. If bid-ask spreads are the highest at the beginning and end of the day as it has been previously documented (e.g., McInish and Wood, 1992), investors attempting to trade during these intervals are more likely to pay higher transaction costs. However, a closer look at the trading prices during the beginning of the trading session indicates that both passive and active foreign investors trade at fair prices during the first hour of market activity, with VWAP of buys and sells indistinguishable from 100 in this interval. Moreover, even investors with more trading activity at the beginning of the day such as domestic passive

16 Most recently, Agudelo et al. (2019) use the VWAP measure to compare performance between different groups of investors in Colombia. The authors, however, do not have investor ID and thus cannot study performance by style, which is the main focus of our paper. 17 The results are similar if we disaggregate the sample into trades of different sizes (results not shown in the table). For smaller trades, foreign investors’ under-performance relative to individuals is smaller but still statistically significant. 18 Consistent with the introduction of the closing auction, Panel C only includes trades after February 9, 2009.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

10

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx Table 6 Value-weighted-average-price by investor type, by Active Share, and for different time intervals. The value-weighted-average-price (VWAP) for each stock s, on day d by investor i is computed as (Bi,s,d ∕As,d ) × 100, where Bi,s,d is the volume-weighted average price for all purchases and sales separately for each investor i, and As,d is the volume-weighted average price for all trades. Panel A presents average V WAP for each investor group and split for buy and sell transactions. For Panel B, foreign investors are classified according to their level of active management: index funds, passive (non-index funds with Active Share below 60%), and active funds (non-index funds with Active Share above 60%). We further disaggregates V WAP for transactions during the continuous trading session and for those in the closing auction and present the results in Panel C. t-statistics are presented in parentheses.

Panel A. VWAP by investor type

Buy: All trades Average of mean VWAP Difference: Foreign - (i)

Sell: All trades Average of mean VWAP Difference: Foreign - (i)

Individuals

Corporations

Pension Funds Mutual Funds

Brokerage FirmsAll Institutions Foreign

99.92 (39.96) 0.11 (13.54)

99.95 (14.82) 0.08 (13.82)

99.90 (12.38) 0.14 (17.27)

99.98 (3.75) 0.06 (8.26)

99.99 (2.81) 0.05 (7.94)

99.96 (13.03) 0.08 (10.27)

100.04 (9.22)

100.06 (50.04) −0.08 (13.24)

100.05 (8.74) −0.07 (5.86)

100.08 (13.50) −0.09 (12.93)

100.06 (12.49) −0.07 (11.91)

100.04 (8.40) −0.06 (8.89)

100.05 (15.06) −0.07 (7.69)

99.98 (3.97)

Panel B. VWAP of foreign investors by level of active management Index

Passive

Active

Buy Average of mean VWAP

100.11(15.42)

100.06(11.97)

100.00(0.33)

Sell Average of mean VWAP

99.91(7.30)

99.97(3.56)

100.01(0.35)

Panel C. VWAP by trading interval Index

Passive

Active

Day session

Closing auction

Day session

Closing auction

Day session

Closing auction

100.05 (3.75)

100.16 (13.68)

100.2 (2.98)

100.16 (18.15)

99.98 (2.91)

100.10 (8.54)

99.93 (4.15)

99.89 (5.83)

99.97 (2.64)

99.91 (7.17)

100.00 (0.67)

99.95 (3.72)

99.98 (2.75)

99.91 (7.45)

99.96 (10.09)

99.89 (10.73)

99.96 (9.32)

99.85 (13.90)

100.03 (13.09)

100.14 (10.94)

100.04 (21.95)

100.19 (18.44)

100.05 (13.37)

100.20 (23.14)

Foreign Investors Buy Average of mean VWAP

Sell Average of mean VWAP

Domestic Investors Buy Average of mean VWAP

Sell Average of mean VWAP

funds —a group that represents 25% of the total traded value in the first hour after market opening– trade at fair prices during this interval.19 4.1. Linear regression analysis So far, we document that foreign investors trade at worse prices relative to both domestic institutions and domestic individuals. Among foreign investors, index funds and those with low Active Share and with more transactions later in the day trade at more unfavorable prices. One limitation to the non-parametric analysis above is that it does not account for differences across investors or in the stocks that they trade. For example, whether greater transaction costs are concentrated in less liquid stocks, or among funds with larger AUM. Furthermore, if foreign investors are less well-informed, their short-term trading underperformance should be explained by firm and stock characteristics that proxy for information asymmetries. For instance,

19

The results are omitted from Table 6 for brevity and are available upon request.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

11

Table 7 Correlation matrix between fund-level variables. The table presents correlation coefficients for the variables of interest. Interval is a dummy variable equal to one for each fund-day if the fund trades (buys and sells separately) during the closing auction. Intensity is the traded value of a fund in each stock-day relative to the total traded volume. The variable Stocks captures the number of different stocks that are traded by each investor during a day (in logs). Trades is the number of transactions for each investor during a day (in logs). Net Flows are daily purchases minus sells normalized by assets under management at the beginning of the month for each investor. Fund turnover is the minimum between yearly purchases and sells normalized by the average of the fund’s assets.

Active Share

Interval buys

Active Share Interval (buys) Interval (sells) Intensity (buys) Intensity (sells) Stocks (buys) Stocks (sells) Trades (buys) Trades (sells) Fund Size Fund Turnover Net Flows

1.00

−0.08 1.00

sells

−0.05

0.16 1.00

Intensity buys

−0.05 −0.01 −0.05 1.00

sells

−0.04

0.00 −0.05 0.49 1.00

Stocks buys

sells

−0.41

0.06 −0.01 0.20 0.18 1.00

−0.36

0.03 0.02 0.16 0.16 0.84 1.00

Trades buys

−0.06 −0.06 −0.08

0.38 0.19 0.24 0.20 1.00

Fund Size

Turnover

Net Flows

−0.27

−0.02 −0.11

−0.09

sells

−0.02 −0.03 −0.12

0.18 0.34 0.20 0.21 0.53 1.00

0.03 −0.09 0.26 0.25 0.61 0.61 0.33 0.33 1.00

0.01 −0.07 −0.07 0.10 0.11 −0.01 −0.01 −0.36 1.00

0.03 0.02 0.05 0.01 0.06 −0.02 0.01 −0.05 −0.01 −0.12 1.00

the disadvantage should be greater in small and opaque firms. In this subsection we use linear regression analysis to further investigate the disadvantage of passively-managed foreign funds. We restrict our analysis to domestic institutions and foreign investors and estimate the VWAPi,s,d for buys and sells separately using the following model: VWAPi,s,d = 𝛼s + 𝜇d + 𝛽 Foreigni + 𝜙1 Indexi + 𝛿1 Foreigni × Indexi

+ 𝜙2 Passivei,d + 𝛿2 Foreigni × Passivei,d + 𝛾 Xi,s,d + 𝜈 Ys,d + 𝜀i,s,d .

(3)

Foreigni is a dummy variable equal to one for foreign investors and zero otherwise. We analyze the role of active management by introducing two dummy variables, Indexi and Passivei,d . Indexi is equal one for index funds and zero otherwise. Passivei,d is set to one for funds that are not index funds and their Active Share is below 60% at the beginning of each month. Xi,s,d contains fund-level information (i.e., fund size and turnover ratio). We are also interested in analyzing whether fund managers are subject to higher transaction costs on days when they are trading more intensively or in trades executed later in the day. To be precise, we use the following variables to study timing, trading intensity, and speed of transactions: (1) (2) (3) (4)

intervali,s,d = dummy variable equal to one for trades executed during the closing batch auction; stocksi,d = log number of different stocks purchased (or sold) by investor i on the day in the same direction of stock s; tradesi,s,d = log number of trades of investor i for a stock-day; intensityi,s,d = buy (or sell) trade value of investor i for a stock-day/total trade value for the stock-day (%).

Equation (3) includes stock-firm controls, Ys,d . We follow Choe et al. (2005) and control for firm size, book-to-market ratio, stock returns, and stock liquidity. Appendix A presents the list of variables used in the regression analysis with their corresponding definition and source. Table 7 reports the correlations between fund variables. In addition to controlling for stock fixed-effects (𝛼 s ), we correct for serial correlation parametrically by including time dummies (𝜇d ) as in Petersen (2009), and calculate standard errors clustered at the fund level. The estimated coefficient 𝛿 1 (𝛿 2 ) is the difference in the average price paid by foreign index (passively-managed foreign) funds relative to active foreign investors. This spread is calculated in excess of the average differences in price ratios between domestic index funds and actively-managed domestic funds. More precisely, 𝛿̂1 = E[VWAP (index, foreign) − VWAP (active, foreign)] − E[VWAP (index, domestic) − VWAP (active, domestic)]. Columns (1) to (3) of Table 8 show the average price of daily buys. According to regression (1), foreign index funds and those with passive management pay 9.7 and 6.7 bps more respectively than actively-managed foreign funds.20 These differences are present after controlling for stock-firm characteristics, fund size, and fund turnover. In other words, the type of stocks that passively-managed foreign funds trade, their size, or turnover ratio do not explain the documented disadvantage in daily buys across groups. In regression (2), we control for the daily trading intensity across funds. The coefficients for the number of transactions, different stocks purchased, and intensity, are all positive and highly significant. These results suggest that fund managers pay higher prices on days when they buy the same stock in multiple transactions (e.g., iceberg trades), buy several stocks during the same day, and represent a large share of the total traded value in each stock. The findings are consistent with the idea that funds

20 For passive funds, the results hold when we lower the threshold of Active Share to 50%. In fact, the underperformance of passively-managed funds appears to be more pronounced, with average buys 7.6 bps more than actively-managed funds. Throughout the paper, we use the 60% cutoff but confirm all our results with the lower threshold.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

12

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx Table 8 VWAP spreads between foreign and domestic institutions: Investor style correlates. This table reports the estimates of daily Value-Weighted-Average-Price (VWAP) according to the model below. The dependent variable in columns (1)–(3) is the VWAP for purchases. VWAP for sells is reported in columns (4)–(6). Foreign, Index, and Passive are dummy variables set to one when the fund is domiciled outside Colombia, is a self-declared index fund, and its Active Share is below 60% respectively. Each model includes controls Xi,s,d with stock and investor characteristics (complete list in Appendix A), time and stock fixed-effects. T-statistics, presented in parenthesis, are calculated from standard errors clustered at the fund level. ∗ ∗∗ , ∗ ∗ , ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level. VWAPi,s,d = 𝛼s + 𝜇d + 𝛽 Foreign i + 𝜙1 Indexi + 𝛿1 Foreigni × Indexi +

𝜙2 Passivei,d + 𝛿2 Foreign i × Passivei,d + 𝛾 Xi,s,d + 𝜈 Ys,d + 𝜀i,s,d . buys

Foreign Index Passive Foreign x Index Foreign x Passive

sells

(1)

(2)

(3)

(4)

(5)

(6)

0.054∗∗ ∗ (5.359) 0.023 (1.544) 0.006 (0.715) 0.097∗∗ ∗ (5.011) 0.067∗∗ ∗ (5.117)

0.042∗∗ ∗ (4.006) 0.021 (1.374) 0.001 (0.156) 0.085∗∗ ∗ (4.422) 0.062∗∗ ∗ (4.180) 0.026∗∗ ∗ (6.358) 0.069∗∗ ∗ (6.631) 0.032∗∗ ∗ (6.263)

0.031∗∗ ∗ (3.216) 0.012 (0.859) 0.000 (0.041) 0.058∗∗ (2.469) 0.026∗∗ (2.091)

−0.063∗∗ ∗

−0.052∗∗ ∗

−0.042∗∗ ∗

(-6.669) −0.032∗∗ (-2.277) −0.008 (-1.056) −0.060∗∗ (-2.233) −0.034∗∗ (-2.390)

(-5.616) −0.032∗∗ (-2.480) −0.007 (-0.851) −0.033∗∗ (-1.975) −0.020∗∗ (-2.183) −0.042∗∗ ∗ (-8.064) −0.015 (-1.557) −0.047∗∗ ∗ (-6.997)

(-4.625) −0.024∗ (-1.941) −0.004 (-0.645) −0.021∗ (-1.809) −0.026∗ (-1.766)

Trades Intensity Stocks Interval

−0.067∗∗ ∗

Foreign x Interval

(-3.869) 0.169∗∗ ∗ (7.199)

Observations R-squared FE Cluster Fund controls Stock controls

464,038 0.026 stock + day Investor Yes Yes

464,038 0.029 stock + day Investor Yes Yes

501,605 0.029 stock + day Investor Yes Yes

0.171∗∗ ∗ (9.748) −0.180∗∗ ∗ (-8.319) 430,355 0.024 stock + day Investor Yes Yes

430,355 0.030 stock + day Investor Yes Yes

465,901 0.030 stock + day Investor Yes Yes

pay more to liquidity providers when they are trading more intensively. Interestingly, even after controlling for observables measures of trading speed and intensity, we find that foreign funds with passive management strategies still pay higher prices on average than actively-managed foreign funds. For instance, while the difference between purchasing prices among foreign index funds and actively-managed foreign funds is lower than the one estimated in model (1), it is still statistically and economically significant at 8.5 bps in the model (2).21 In model (3), we estimate V WAP by splitting the sample according to the time interval of each trade — continuous trading session and closing auction. Consistent with the aggregate differences reported in Panel C of Table 6, foreign funds buy stocks at unfavorable prices at the closing auction. These late trades explain almost half the disadvantage of foreign index funds (i.e., the spread relative to active funds decreases from 9.7 bps to 5.8 bps). In Table 8, we also investigate the determinants of average sell prices. As expected, the coefficients in columns (4)–(6) have the opposite sign from the coefficients of buys. For example, foreign index funds receive 6.0 bps less for stock sales than foreign active funds. This spread is smaller at 3.3 bps but statistically significant when the controls for trading intensity are introduced (model 5). While the analysis in this section focuses on the inferior trading performance of passively-managed foreign funds vis-ávis their more actively-managed counterparts, we can also use the estimates from equation (3) to examine differences across actively-managed funds in our sample. For example, relative to domestic funds, actively-managed foreign funds pay 5.4 bps (model 1) for buys and receive 6.3 bps for sells (model 4). These findings should be interpreted carefully. As we report in Panel A of Table 6, actively-managed foreign funds trade at the average daily price, with buys and sells indistinguishable from 100. They only appear at a disadvantage relative to domestic funds because local institutional investors display better than average

21 We also control for average trade size in unreported results. While larger trades are related to higher VWAP, the estimated gap in trading performance between domestic and foreign investors remains unchanged.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

13

performance by taking the opposite side of the inferior trades of index and passively-managed foreign funds. In summary, the average disadvantage of foreign investors is more pronounced among those with passive management strategies (i.e., index funds and funds with low Active Share). It appears that these funds are paying a price for a trading style that differs from the style of other funds. In particular, managers of passively-managed foreign funds trade at worse prices at the end of the day and when they are trading more intensively. However, even after controlling for trading interval, including measures of intensity, using stock fixed effects and introducing variables that proxy for information asymmetries, there are some remaining differences in average prices by level of active management. In other words, our measures of trading activity and proxies of information cannot fully explain the observed disadvantage of foreign funds, especially for those that are passive. For instance, while the difference in transaction prices across groups might result from price pressure exerted by passivelymanaged foreign funds, it is possible that these funds trade after prices have moved against them due to intra-day momentum strategies. Alternatively, if foreign investors are better informed, their trades would have a greater permanent price impact. We come back to these potential explanations in Section 5 where we study the relation between market returns and equity flows disaggregated by investor domicile and by management style.

4.2. Risk-adjusted returns Our findings suggest that passively-managed foreign funds trade at unfavorable prices at daily and intra-day frequencies. Inferior trading at the stock level, however, does not imply poor portfolio performance if the stocks that a fund buys display superior returns during the fund’s holding period, or if hedging strategies across stocks generate superior risk-return trade-offs. We complement our earlier findings by examining total portfolio returns by level of active management. The objective is to test whether index and passively-managed foreign funds, in addition to trading at worse prices, generate lower returns in their holdings of Colombian stocks. We use monthly holdings in Colombian stocks and net purchases to calculate the monthly gross rate of returns for each fund. More precisely, we calculate the rate of return of fund i in month t (Ri,t ) as Ri,t =

VFi,t −NFi,t VFi,t−1

− 1, where V F is the value of the fund,

and net flows NF include the net value of all stocks buys and sells during the month, as well as dividend payments. We use the cross-sectional variation to see how returns vary between domestic and foreign funds, and between activelymanaged versus passively-managed funds. We adjust for heterogeneity in risk taking and in style by introducing various performance benchmarks that account for the possibility that funds load differently on small-cap stocks, value stocks, and price momentum strategies. To be precise, we adjust monthly fund returns in three ways: (1) we calculate market-adjusted returns by subtracting the returns of a market index; (2) we adjust returns using the capital asset pricing model (CAPM); and (iii) using the Carhart four-factor model (Carhart, 1997). The portfolios that make up our performance benchmarks are the return on the market index in excess of the one-month Colombian T-bill rate (MARKET),22 the returns to the Fama and French (1993) SMB (small stocks minus large stocks) and HML (high book-to-market stocks minus low book-to-market stocks) portfolios, and the returns-to-price momentum portfolio WML (winners minus losers, constructed based on a twelve-month formation period and a one-month holding period). Since we are interested in the relation between management style and performance, we sort domestic and foreign funds at the beginning of each month into our three groups of analysis: index, passive, and active funds. We then track these six portfolios for one month and use the entire time series of their monthly returns to calculate the loadings to the various factors for each of these portfolios. Table 9 reports the average of monthly market-adjusted returns and the loadings of the domestic and foreign funds for each group of active management. According to Panel A, index and passively-managed foreign funds display lower monthly market-adjusted returns and smaller alphas than actively-managed foreign funds. For example, while foreign active funds have a monthly estimated alpha of 0.72% in the four-factor model, the corresponding point estimate for the alpha of foreign index funds is 0.15% and is statistically indistinguishable from zero. In other words, actively-managed foreign funds appear to deliver higher risk-adjusted returns relative to foreign funds with passive management strategies. Panel B of Table 9 reports that all three groups of domestic institutional investors display positive risk-adjusted returns. This finding raises at least two concerns related to the validity of the risk-adjustment model. First, it appears that our investor groups have either no excess performance relative to the market (foreign index and passive) or their returns are above the market (foreign active and domestic institutions). By construction, if some investors are overperforming relative to the market, some investors must be underperforming. As it turns out, domestic individuals yield below market returns when their performance is measured at monthly or yearly frequencies. The evidence that medium- and long-term performance of domestic institutions is above that of foreign investors in Colombia, and that foreigners in turn outperform domestic individuals was first documented by Agudelo et al. (2019). Here, we extend this finding by disaggregating performance by type of active management. Second, the fact that even domestic index funds display positive risk-adjusted returns would suggest that some of these funds are tracking benchmarks that are different than the Colombian market index. Also, index funds, whether domestic or foreign, often hold cash reserves (up to 10% of AUM). In turn, some differences in performance relative to their benchmarks

22 We use two other measures for the risk-free rate: the monthly return on U.S. T-Bills in Colombian pesos and the Colombian deposit rate. Our results are unchanged when using these proxies for the risk-free rate. We omit these results for brevity.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

14

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

Table 9 Performance benchmarks. This table reports the loadings of the portfolios by active management group: index funds, passive (non-index funds with Active Share below 60%), and active funds (non-index funds with Active Share above 60%). Panel A reports loadings of foreign investors and Panel B for domestic institutions. MARKET is the return of the Colombian market index in excess of the one-month Colombian T-bill rate. SMB is the return on a portfolio of small stocks minus large stocks. HML is the return on a portfolio long on high book-to-market stocks and short low book-to-market stocks. WML is the return on a portfolio long on stocks that are past-12-month winners and short on those that are past-12-month losers. ∗ ∗∗ , ∗ ∗ , ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level.

Active Share

Market-adj (average)

CAPM

Carhart 4-Factor

alpha

MARKET

alpha

MARKET

SMB

HML

WML

0.17% (1.15) 0.15% (1.07) 0.79%∗∗ ∗ (3.01)

0.87∗∗ ∗ (25.03) 0.90∗∗ ∗ (31.92) 0.74∗∗ ∗ (18.23)

0.15% (0.99) 0.12% (1.02) 0.72%∗∗ ∗ (2.94)

0.82∗∗ ∗ (21.40) 0.86∗∗ ∗ (28.05) 0.72∗∗ ∗ (15.05)

−0.15∗ ∗ (2.18) −0.11∗ (1.85) −0.34∗ ∗ ∗ (4.00)

−0.10 (1.37) −0.08 (1.58) 0.02 (0.03)

0.01 (0.01) 0.05 (0.92) −0.03 (0.38)

0.91∗∗ ∗ (31.68) 0.90∗∗ ∗ (31.81) 0.81∗∗ ∗ (22.50)

0.49%∗∗ ∗ (3.02) 0.44%∗∗ ∗ (3.16) 0.62%∗∗ ∗ (3.24)

0.90∗∗∗ (27.48) 0.89∗∗ ∗ (27.43) 0.71∗∗ ∗ (19.21)

−0.09 (1.47) −0.20∗ ∗ ∗ (3.30) −0.18∗ ∗ ∗ (2.59)

−0.03 (0.53) 0.03 (0.54) −0.03 (0.45)

0.07 (1.31) 0.01 (0.15) 0.06 (0.97)

Panel A. Foreign Investors Index Passive Active

0.20% (1.17) 0.18% (1.18) 0.85%∗∗ ∗ (3.45)

Panel B. Domestic Institutional Investors Index Passive Active

0.59%∗∗ ∗ (3.93) 0.55%∗∗ ∗ (3.50) 0.70%∗∗ ∗ (3.49)

0.53%∗∗ ∗ (3.83) 0.49%∗∗ ∗ (3.30) 0.66%∗∗ ∗ (3.41)

might result if the fund managers possess some stock picking skills or market timing ability. Since we do not observe funds’ identities, we cannot directly calculate the excess performance relative to each fund self-declared benchmark (which would be the ideal setting to test for relative performance). Instead, we test the null hypothesis that the performance of foreign index and passively-managed funds is similar to the returns of other foreign and local funds, when returns are adjusted for exposure to different risks. So far, the evidence seems to suggest that we can reject this null hypothesis. That is, in addition to trading at worse prices, passively-managed foreign funds display lower returns than actively-managed foreign and local institutional investors. For example, the difference between the monthly market-adjusted returns of domestic and foreign index funds is economically important at 0.39%, or 4.78% in yearly returns.23 It is possible that Active Share is correlated with other fund characteristics that are driving performance. For example, fund size might erode performance because of trading costs associated with liquidity and price impact (Chen et al., 2004b). Since passive funds are larger (see Table 5), the reported under-performance of passively-managed foreign funds might be a consequence of their size rather than their level of active management. To deal with the correlation between Active Share with other fund characteristics, we analyze the effect of past Active Share on performance in a regression framework proposed by Fama and James (1973), where we control for the effects of other fund characteristics on performance. We use the following specification: ADJRETi,t = 𝜇 + 𝛽 Foreigni + 𝜙1 Indexi + 𝛿1 Foreigni × Indexi + 𝜙2 Passivei,t−1

+ 𝛿2 Foreigni × Passivei,t−1 + 𝛾 Xi,t−1 + 𝜀i,t fori = 1, … , N

(4)

where ADJRETi,t is the return of fund i in month t adjusted by various performance benchmarks and Xi,t−1 is a set of control variables that includes fund size, turnover, and net flows. Foreign, Index, and Passive are dummy variables defined in the previous section. We take the estimates from these cross-sectional monthly regressions and follow Fama and James (1973) in taking their time series means and standard deviations to form our overall estimates. In Table 10, Panel A presents the estimated coefficients of equation (4) and Panel B shows the estimated differences in monthly performance across groups. For example, the average difference in risk-adjusted returns between actively-managed

̂1 + 𝛿̂1 ). We confirm foreign funds and foreign index funds is given by: E[ADJRET (active, foreign) − ADJRET (index, foreign)] = −(𝜙 our finding that foreign index and passively-managed foreign funds underperform foreign active funds. The difference in riskadjusted returns is both statistically and economically significant. Using the most conservative estimates, namely, those from the four-factor model, actively-managed foreign funds outperform foreign index funds by 2.30 percentage points per year (0.19 per month) and passively-managed foreign funds by 2.62 percentage points per year (0.22 per month). Although these are gross returns (i.e., they do not take into account management fees), the difference is above 0.8%, which is the average expense ratio that U.S. active funds charge in excess of passive funds.24 The results here only refer to the performance of the share invested in Colombian equity. However, the findings are important because they suggest that the inferior trading of passively-managed foreign funds cannot be explained by hedging or investThis difference is statistically significant at the 5% confidence level (the covariance between the two series of adjusted returns is −4.38 × 10−6 ). According to Thomson Reuters Lipper, the average expense ratio of actively-managed equity funds in the U.S. is 1.4%, while the average expense ratio of passively-managed funds is 0.6%. 23 24

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

15

Table 10 Regression of fund performance by level of active management. This table reports the Fama and James (1973) estimates of monthly fund returns regressed on fund characteristics lagged one month (model specified below). Monthly returns (ADJRET) are adjusted using the market model, the CAPM, and the four-factor model. Foreign, Index, and Passive are dummy variables set to one when the fund is domiciled outside Colombia, is a self-declared index fund, and its Active Share is below 60% respectively. Other control variables are defined in Appendix A. Panel A presents estimation results and Panel B displays the estimated differences across investor groups from the regression. The t-statistics are adjusted for serial correlation using the Newey and West (1987) lags of order three and are shown in parenthesis. ∗∗∗ , ∗∗ , ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level. ADJRETi,t = 𝜇 + 𝛽 Foreign i + 𝜙1 Indexi + 𝛿1 Foreign i × Indexi

+𝜙2 Passivei,t−1 + 𝛿2 Foreign i × Passivei,t−1 + 𝛾 Xi,t−1 + 𝜀i,t . Market-adj

CAPM

4-Factor

0.139 (1.120) 0.142 (0.963) −0.009 (-0.077) −0.412∗∗ (-2.232) −0.296∗∗ (-2.252) 0.013 (0.780) 0.013 (0.652) −0.083∗∗ ∗ (-3.027) 159,357

0.129 (1.006) 0.199 (1.353) 0.067 (0.600) −0.436∗ ∗ (-2.329) −0.322∗ ∗ (-2.294) 0.013 (0.833) 0.017 (0.822) −0.077∗ ∗∗ (-2.786) 156,426

0.119 (0.972) 0.213 (1.393) 0.070 (0.614) −0.403∗∗ (-2.143) −0.287∗∗ (-2.091) 0.013 (0.833) 0.017 (0.822) −0.077∗∗ ∗ (-2.786) 156,426

0.139 (1.120) −0.272∗∗ (1.978) −0.157∗ (1.823) 0.269∗ (1.691) 0.305∗∗ (2.271) −0.142 (0.963)

0.129 (1.006) −0.307∗ ∗ (1.986) −0.193∗ ∗ (2.153) 0.237∗ (1.687) 0.254∗∗ (2.049) −0.199 (1.353)

0.119 (0.972) −0.283∗ (1.698) −0.168∗ (1.842) 0.190∗ (1.650) 0.216∗ (1.693) −0.213 (1.393)

Panel A. Regression Results Foreign Index Passive Foreign x Index Foreign x Passive Fund Size Turnover Net Flows Observations Panel B. Differences in performance between groups Active Foreign - Active Domestic = 𝛽 Foreign Index - Domestic Index = 𝛽 + 𝛿 1 Passive Foreign - Passive Domestic = 𝛽 + 𝛿 2 Active Foreign - Foreign Index = −(𝜙1 + 𝛿 1 ) Active Foreign - Passive Foreign = −(𝜙2 + 𝛿 2 ) Active Domestic Active - Domestic Index = −𝜙1

ment strategies that are delivering higher overall returns within Colombian stocks.25 Moreover, when compared to domestic funds with similar management strategies, foreign index funds and passively-managed foreign funds deliver lower risk-adjusted returns, between 1.94 and 3.78 percentage point per year (0.157 and 0.307 per month). The returns of actively-managed foreign funds, on the contrary, are indistinguishable from those of actively-managed domestic funds. To summarize, the under-performance of foreign investors in the Colombian stock market is attributable to passivelymanaged funds. These funds trade at worse prices and deliver lower risk-adjusted returns relative to actively-managed funds. While the under-performance can be explained in part by mechanical strategies, such as buying and selling multiple stocks in the portfolio at the same time near closing prices, the fact that domestic passively-managed funds have superior performance suggests that there are other fund characteristics besides passive management that might explain the under-performance of foreign funds. In the next section, we evaluate alternative explanations for the trading underperformance of passively-managed foreign funds. 5. The price impact of investors’ flows Subsection 4.1 showed that controlling for trading intensity and for firm and stock characteristics does not explain the disadvantage in short-term trading by passively-managed foreign funds. There are at least three potential explanations for their trading underperformance. First, if these funds have superior information, their transactions would signal that information,

25 An alternative explanation that we cannot rule out is that foreign index funds trade at worse prices in Colombia but at the same time they are better at cross-country diversification, which then improves their overall risk-return profile of these funds. More precisely, we cannot test the hypothesis that despite trading at worse prices, passive foreign investors are better at cross-country hedging than other foreign investors.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

16

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

shifting expectations, and thereby generating a permanent price impact. Second, with incomplete risk-sharing, large trades by this group would temporarily disturb the position of liquidity providers, which requires the buyer to pay compensation in the form of higher prices. Third, if foreign funds follow intra-day momentum strategies, they might choose to trade after prices have moved. In order to evaluate the competing hypothesis behind the poor trading performance of passively-managed foreign funds, in this section we examine the joint dynamics of local market returns and net equity flows for each investor group. 5.1. The dynamics of equity flows and market returns To investigate these issues, we use a vector autoregression (VAR) framework. Our strategy follows Griffin et al. (2004), with the key difference that instead of tracking total daily equity flows by foreign investors, we disaggregate flows for each group of active management. Flows are defined as the value of all equity purchases by each investor group i minus all equity sales, scaled by the previous day’s market capitalization, fi,t = 100(buysi,t − sellsi,t )∕mktcapt−1 .26 Table 11 displays the VAR regression results for both flows and domestic market returns.27 According to Panel A, flows are strongly related to their past values. For example, a one standard deviation increase in the previous day’s flows of foreign index funds leads to a 0.18 standard deviation increase in today’s flows. Flow persistence, however, is observed for both domestic and foreign funds, as well as for passive and actively-managed funds. The effects from the previous day’s returns on flows vary across investors. The evidence suggests that foreign investors buy following high previous-day stock returns but respond little several days later. An increase of one standard deviation in domestic market returns is followed by next-day increases in foreign flows between 0.059 and 0.098 standard deviations. Interestingly, domestic index funds also appear to be chasing local market returns. It appears that momentum investing is not an exclusive behavior of passively-managed foreign funds. Active domestic investors, on the other hand, display contrarian strategies — buying after negative previous-day returns — a behavior that is consistent with evidence on domestic investors in Korea and Finland (Choe et al., 1999; Grinblatt and Keloharju, 2000). We next use the return equation to examine the relation between current domestic market returns and past trading activity. In Panel B of Table 11, Lag 1 flows of domestic investors (passive and active) are significant predictors of returns, indicating that these investors are buying before the domestic market increases. In contrast, the previous day’s equity flows of foreign index and passive funds are negatively related to today’s domestic stock returns.28 If foreigners are paying higher prices for their purchases due to liquidity frictions, this price impact is likely transitory and should reverse. To differentiate between transitory and permanent price effects, we include in the return equation the contemporaneous flows for each investor group. Panel C of Table 11 shows that the contemporaneous flows are positive and highly significant for five of the six investor groups. Importantly, the lagged flows for index and passive foreign funds are still negative and significant. The complete dynamics from the effects of flows on domestic asset prices can be read as follows: A one standard deviation increase in daily flows of foreign index funds is correlated with a 0.173 standard deviation increase in domestic stock prices on the same day and is followed by a 0.103 decline in prices the next day. The difference between the contemporaneous increase in prices and the next day reversal (0.173 − 0.103 = 0.07 of one standard deviation in returns) can be interpreted as the permanent price effect, and the reversal can be attributed to the temporary distortion due to liquidity frictions. Since net flows are normalized by market capitalization, our findings should be symmetric between investors who demand liquidity and those that provide liquidity. In our case, actively-managed domestic funds appear to be compensated for providing liquidity: their sells (negative flows of one standard deviation) are associated with an increase of 0.107 standard deviation in prices, which is the temporary price distortion from index funds. Interestingly, Panel C also documents that the contemporary price impact of large stock transactions by domestic index funds is also reversed, which further provides support for the view that liquidity frictions are driving our results. Also, we do not find evidence that the managers of passively-managed foreign funds are better informed than other domestic institutions. The clear message from our analysis is that passively-managed foreign funds are experiencing higher transactions costs in the form of a larger temporary price impact. 5.2. The response of flows to international markets To examine the drivers of foreign investors demands, we study the extent to which these funds buy or sell Colombian stocks when asset prices are rising or falling in other countries. Since foreign funds are likely to exhibit global investment strategies, it is possible that their equity flows to Colombia are more responsive to international news than Colombian funds with a more domestic focus (Griffin et al., 2004).29 To examine the importance of global and regional indices in explaining flow dynamics,

26

We set observations above the 99th percentile of the daily net flow distribution equal to the 99th percentile point. For most investor groups, the Hannan-Quinn information criterion selects the optimal lag length at four lags. We model the system with four lags in each variable for all investor groups to make the analysis homogeneous across management style. For brevity, the table only presents the first two lags for each variable. 28 The adjusted R2 ’s in the return equations are less than 0.04 in all the specifications. Nevertheless, we wish to further understand the relation between flows and next-day stock returns. 29 For instance, according to Factset and Lipper, among 294 mutual funds with at least one equity investment in Colombia during 2015, 53% use an emerging market index as their benchmark, 20% a world index, and 18% use an index that tracked the returns of several markets in Latin America. 27

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

17

Table 11 Vector autoregression returns of returns and net flows by investor type. This table reports results from the bivariate vector autoregression (VAR) specified below with four lags for each endogenous variable. rt is the daily return of the Colombian stock market index, fi,t is the daily net equity flows (buy - sell value) originated by investor group i scaled by the previous-day market capitalization, the 𝛼 ′s are constant intercept terms, b(L) denotes a polynomial in the lag operator L, and 𝜀’s are zero-mean disturbance terms that are assumed to be intertemporally uncorrelated. Returns and flows are standardized. The VAR is estimated separately for each fund type by OLS. Panels A and B report coefficients estimates, their t-statistics, and adjusted R2 for the flow and return equation respectively. Panel C reports the return equation results for a structural VAR with contemporaneous flows in the return equation. ∗∗ ∗ , ∗∗ , ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level. The VAR equations are as follows:

[ ] rt

fi,t

[

=

𝛼i,r 𝛼i,t

]

[

+

][

b11 (L)

b12 (L)

b21 (L)

b22 (L)

rt−1

fi,t−1

]

[

+

𝜀ri,t 𝜀

f i,t

]

.

Foreign Investors Index

Domestic Investors Passive

Active

Index

Passive

Active

0.148∗∗ ∗ (5.431) 0.140∗∗∗ (5.077) 0.059∗∗ (2.223) 0.036 (1.327) .086

0.231∗∗ ∗ (8.648) 0.121∗∗ ∗ (4.433) 0.048∗ (1.956) 0.062∗∗ (2.502) .141

0.132∗∗ ∗ (4.924) 0.088∗∗ ∗ (3.245) 0.113∗∗ ∗ (4.276) 0.031 (1.144) .069

0.243∗∗ ∗ (9.043) 0.100∗∗ ∗ (3.630) −0.013 (-0.487) −0.040 (-1.537) .126

0.155∗∗ ∗ (5.734) 0.138∗∗ ∗ (5.014) −0.098∗ ∗∗ (-3.823) −0.053∗ ∗ (-2.082) .099

−0.063∗∗

−0.053∗∗

(-2.210) −0.029 (-0.980) 0.131∗∗∗ (4.863) 0.046∗ (1.670) .029

(-1.977) −0.003 (-0.103) 0.122∗∗ ∗ (4.503) 0.042 (1.511) .023

0.013 (0.460) 0.059∗∗ (1.966) 0.123∗∗ ∗ (4.541) 0.053∗∗ (1.961) .036

−0.033 (-1.223) −0.051∗ (-1.840) 0.098∗∗ ∗ (3.623) 0.034 (1.263) .017

0.055∗∗ (1.960) −0.013 (-0.442) 0.101∗∗ ∗ (3.728) 0.048∗ (1.743) .027

0.050∗ (1.782) 0.019 (0.661) 0.099∗∗ ∗ (3.718) 0.062∗∗ (2.340) .025

0.086∗∗ ∗ (2.959) −0.007 (-0.236) 0.054∗ (1.813) 0.115∗∗ ∗ (4.249) 0.044 (1.603) .043

0.127∗∗ ∗ (4.671) −0.063∗∗ (-2.280) −0.063∗∗ (-2.267) 0.082∗∗ ∗ (3.021) 0.023 (0.850) .033

0.060∗∗ (2.140) 0.032 (1.121) −0.024 (-0.835) 0.090∗∗ ∗ (3.312) 0.056∗∗ (2.042) .028

−0.107∗ ∗∗

Panel A. Flow equations Net flows Lag 1 Net flows Lag 2 Returns Lag 1 Returns Lag 2 Adj. R2

0.182∗∗∗ (6.727) 0.099∗∗∗ (3.558) 0.098∗∗∗ (3.813) 0.024 (0.923) .102

Panel B. Return equations Net flows Lag 1 Net flows Lag 2 Returns Lag 1 Returns Lag 2 Adj. R2

Panel C. Return equations with contemporaneous flows Net flows Lag 0 Net flows Lag 1 Net flows Lag 2 Returns Lag 1 Returns Lag 2 Adj. R2

0.173∗∗∗ (6.193) −0.103∗∗ ∗ (-3.552) 0.000 (-1.318) 0.116∗∗∗ (4.303) 0.042 (1.534) .057

0.130∗∗∗ (4.725) −0.051∗ (-1.813) −0.022 (-0.759) 0.122∗∗ ∗ (4.481) 0.051∗ (1.850) .04

(-3.823) 0.071∗∗ (2.473) 0.038 (1.318) 0.087∗∗ ∗ (3.254) 0.061∗∗ (2.301) .036

we estimate a structural VAR system where net equity flows for each investor group and the Colombian stock index depend on their lagged values as well as those of developed, emerging, and Latin America market indices. In this structural VAR, the returns of international indices are assumed to be exogenous variables, and are calculated from daily changes in the following indices: S&P Developed BMI, S&P Emerging BMI, and the S&P Latin America BMI.30 Table 12 displays the results for the flow regressions. Panel A shows the results when only lagged international returns are included in the VAR. Panel B shows the results of the VAR specification when we include the contemporaneous returns of the three international indices. The evidence suggests that among foreign investors, equity demands by index and passive funds display a strong contemporaneous correlation with the returns of stocks in other emerging and regional markets. The economic magnitude of this effect is substantial: a one standard deviation shock to the Latin America index returns coincides with an increase between 0.080 and 0.137 of the same-day flows of index and passively-managed foreign funds respectively. Conversely, a shock to the developed market index returns leads to a same-day decrease in flows for passively-managed foreign funds in Colombia. In other words, stock purchases by foreign funds with passive strategies coincide with days of high returns in emerging and Latin American equity markets, while sells take place during days of high-returns in developed markets. Activelymanaged foreign funds, on the contrary, do not appear to trade in the direction of international markets returns. Table 12 also documents the behavior of domestic funds. Some of these domestic institutional investors seem to be trading in the opposite

30 Returns are calculated in local currency. We also reexamine our main results using the S&P 500 instead of the S&P Developed BMI to capture developedmarket returns and obtain similar findings.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

18

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx Table 12 VAR: Flows by investor group with international index returns. This table reports results from a bivariate VAR of flows and returns. Returns and flows are standardized. Daily returns on international market indices, which are considered to be exogenously determined, are included in the VAR. Four lags are used for all endogenous and exogenous variables, but only the first two lags are reported. All returns are calculated in local currency. The VAR is estimated separately for each investor group by OLS. Panel A report coefficient estimates, t-statistics, and R2 for the flow equation from a standard VAR with no contemporaneous variables in either equation. Panel B reports estimates when the contemporaneous international index returns are included in the VAR equation. ∗ ∗∗ , ∗∗ , ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level.

Foreign Investors

Domestic Investors

Index

Passive

Active

Index

Passive

Active

0.178∗∗ ∗ (6.480) 0.098∗∗ ∗ (3.467) 0.067∗∗ (2.259) 0.008 (0.275) −0.025 (-0.529) 0.027 (0.554) −0.022 (-0.415) −0.037 (-0.676) 0.105∗∗ (2.130) 0.050 (0.976) .111

0.152∗∗ ∗ (5.542) 0.134∗∗ ∗ (4.787) 0.068∗∗ (2.222) 0.008 (0.247) 0.028 (0.592) 0.023 (0.475) −0.108∗∗ (-1.986) −0.011 (-0.193) 0.028 (0.541) 0.048 (0.909) .103

0.230∗∗ ∗ (8.556) 0.123∗∗∗ (4.470) 0.036 (1.262) 0.055∗ (1.942) −0.014 (-0.301) −0.057 (-1.240) −0.014 (-0.266) −0.038 (-0.727) 0.049 (1.006) 0.097∗ (1.933) .147

0.133∗∗ ∗ (4.903) 0.088∗∗∗ (3.200) 0.103∗∗∗ (3.352) 0.039 (1.261) 0.012 (0.246) −0.037 (-0.750) −0.063 (-1.146) 0.019 (0.333) 0.061 (1.217) 0.016 (0.317) .072

0.240∗∗ ∗ (8.767) 0.104∗∗ ∗ (3.694) −0.003 (-0.115) −0.040 (-1.325) −0.069 (-1.474) −0.062 (-1.277) 0.084 (1.531) 0.071 (1.260) −0.038 (-0.763) −0.027 (-0.525) .136

0.151∗∗ ∗ (5.527) 0.131∗∗ ∗ (4.730) −0.129∗∗ ∗ (-4.407) −0.028 (-0.957) 0.043 (0.908) 0.019 (0.391) −0.020 (-0.365) 0.008 (0.142) 0.064 (1.278) −0.053 (-1.021) .122

0.132∗∗ ∗ (4.926) 0.094∗∗∗ (3.452) 0.107∗∗∗ (3.511) 0.039 (1.275) 0.110∗∗ (2.311) −0.008 (-0.166) −0.004 (-0.077) −0.049 (-0.882) −0.023 (-0.477) 0.075 (1.470) .079

0.242∗∗ ∗ (9.004) 0.101∗∗ ∗ (3.652) −0.001 (-0.027) −0.037 (-1.245) 0.095∗∗ (2.046) −0.034 (-0.710) −0.161∗∗ ∗ (-3.008) 0.039 (0.700) −0.015 (-0.317) −0.010 (-0.189) .146

0.154∗∗ ∗ (5.711) 0.139∗∗ ∗ (5.074) −0.131∗∗ ∗ (-4.514) −0.025 (-0.875) −0.010 (-0.211) 0.061 (1.278) −0.066 (-1.232) −0.039 (-0.708) 0.022 (0.460) 0.072 (1.406) .122

Panel A. Flow equations Net flows Lag 1 Net flows Lag 2 Returns Lag 1 Returns Lag 2 Developed Lag 1 Developed Lag 2 Emerging Lag 1 Emerging Lag 2 Latin America Lag 1 Latin America Lag 2 R2

Panel B. Flow equations with contemporaneous international returns Net flows Lag 1 Net flows Lag 2 Returns Lag 1 Returns Lag 2 Developed Lag 0 Developed Lag 1 Emerging Lag 0 Emerging Lag 1 Latin America Lag 0 Latin America Lag 1 R2

0.175∗∗ ∗ (6.457) 0.102∗∗ ∗ (3.654) 0.065∗∗ (2.236) 0.004 (0.120) −0.155∗ ∗ ∗ (-3.365) −0.045 (-0.949) 0.109∗∗ (2.052) 0.009 (0.171) 0.080∗ (1.683) 0.077 (1.546) .119

0.153∗∗ ∗ (5.666) 0.135∗∗ ∗ (4.915) 0.067∗∗ (2.213) 0.002 (0.066) −0.141∗∗ ∗ (-3.005) 0.007 (0.152) 0.096∗ (1.801) −0.074 (-1.349) 0.137∗∗ ∗ (2.794) 0.005 (0.096) .114

0.232∗∗∗ (8.681) 0.122∗∗∗ (4.447) 0.037 (1.320) 0.057∗∗ (2.007) 0.035 (0.782) −0.019 (-0.417) −0.015 (-0.298) −0.012 (-0.225) 0.007 (0.148) 0.056 (1.135) .147

direction of foreign funds. For example, passive domestic investors buy local stocks on days with high developed market returns and sell when emerging market returns are positive.31 We confirm the overall dynamics by looking at impulse response functions for investor flows from an orthogonalized VAR where flows are ordered before domestic returns. This structure implies that flows are allowed to contemporaneously affect domestic market returns but the domestic returns are not allowed to contemporaneously affect flows. Also, the returns of international indices are assumed exogenous and are allowed to affect both flows and domestic market returns. Impulse response graphs are shown in Fig. 2 for each investor group. Shocks to developed market returns lead to large contemporaneous outflows

31 To estimate the variation in flows that can be explained by the different equity indices, we compare the adjusted R2 s when only local index returns are included (as in Panel A of Table 11) with the adjusted R2 s when international indices are included. We find that the adjusted R2 for flows of passively-managed foreign funds with only lagged flows and domestic market returns is only 0.086, but the explanatory power increases by 19.7% to 0.103 with the inclusion of lagged international index returns and increases an additional 10.6% to 0.114 with the inclusion of contemporaneous international returns. These foreign returns appear to be economically important in determining the variation in flows of funds with passive strategies.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

19

Fig. 2. Response of flows to return shocks. The figure shows impulse response functions of net flows to a one standard deviation shock in developed, emerging, and Latin America market return. Responses are expressed in standard deviation units. The time scale is in days. Results are based on the VAR specified in Table 12. The VAR is estimated separately for each investor group, with four lags for each endogenous and exogenous variable. Returns are calculated in local currency. The shaded are represents the 95% confidence intervals.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

20

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

by foreign index and passive funds. Conversely, shocks to emerging and Latin America market returns lead to large same-day inflows: a one standard deviation increase in emerging markets returns leads to a 0.10 standard deviation increase in net flows in the same day by passively-managed foreign funds. The effect on flows from changes in international indices, however, seems short-lived. Overall, the evidence suggests that passively-managed foreign funds have a strong same-day response to global and regional returns. While selling (buying) assets from one country when asset prices fall (rise) in another has been extensively documented for investors and managers in emerging markets [e.g., Kaminsky et al. (2004) using quarterly portfolio data], our evidence suggests that such behavior might be more pronounced or exclusive to passively-managed foreign funds. There are at least two explanations for this type of trading, commonly known as “contagion trading.” First, investors might optimally use asset prices signals in a country/region to update their expectations about returns in their target countries.32 Furthermore, investors with less country-specific information might overweight public signals and trade more intensively on unexpected changes in global asset prices than informed investors in the host country (Brennan and Cao, 1997; Calvo and Mendoza, 2000). Second, foreign investors might engage in cross-market hedging (Kodres and Pritsker, 2002). For instance, a positive shock to asset prices in one country can lead investors to buy that country’s stocks, increasing their exposure to the idiosyncratic factor of the country. Investors then hedge this new position by buying the assets of a second country with non-related fundamentals. Given our single-country data, we cannot test whether the local response to international returns is an efficient strategy from updating expectations, cross-hedging, or from the behavior or less well-informed investors.33 Our results, however, indicate that the intensity of the trades form passively-managed funds, often driven by global returns, combined with illiquidity frictions in the host country, lead to significant transaction costs — via larger temporary price impact — and lower portfolio returns. 5.3. Price pressure and momentum trading In the previous sections, we study flows at the fund-day level. In order to examine the role of herding behavior and contrast price effects from the alternative hypothesis that worse performance results from momentum strategies, we focus on fund-stock level demands for day and intra-day frequencies. We calculate the momentum measure of Grinblatt (1995) using daily trades on individual stocks as follows: Mi,k =

D S 1∑∑ T d=1 s=1

(

Qi,s,d − Qi,s,d−1 Q i,s,d

) Rs,d−k ,

(5)

where Qi,s,d are holdings by fund i of stock s at day d, Q i,s,d is defined as (Qi,s,d + Qi,s,d−1 )∕2, and Rs,d−k is the return of stock s from d − k − 1 to d − k. When k = 1, the measure captures the lagged response of trades to returns (“lag-1 momentum” or L1M). When k = 0, this measure captures the contemporaneous relation between trades and returns (“lag-0 momentum” or L0M). We calculate L1M and L0M by averaging across investors in the same group of active management. We find that buys and sells by index and passively-managed foreign funds are systematically related to stock returns in the same day (continuous trading in Table 13). For example, the average of L0M for passively-managed foreign funds is 0.16 (t-statistic of 5.17), which is over five times greater than the correlation between flows and lag returns for this group. Among foreign investors, we find that 79% of index funds and 72% of passive funds buy and sell in the direction of the same-day returns. Active foreign funds, on the other hand, do not appear to trade in the direction of daily stock returns. Meanwhile, domestic institutions trade in the opposite direction of returns, with L0M negative for all three groups of management style. It is possible that passively-managed foreign funds are following intra-day momentum strategies. For example, managers of these funds might buy (sell) intensively at the closing batch after prices have already gone up (down) during the continuous trading session. In this scenario, the contemporaneous correlation between daily returns and demands would not result from price pressure, but from intra-day momentum behavior that is not captured in the estimates using daily data. To account for this possibility, we calculate the measure of momentum by fund and trading interval. To be precise, we split the daily trading session in two subperiods: the continuous trading session and the closing auction. Returns in the closing auction are calculated using the adjudication price (which is also the closing price of the day) and the price of the last trade in each stock before the start of the auction. Returns during the continuous trading session are based on the price before the start of the closing batch auction and the closing price of the previous day. According to Table 13, the L0M measure among passively-managed foreign funds and index funds is positive and statistically significant in both the continuous trading session and the closing auction. Moreover, a significant percentage of index funds trade in the direction of returns in both intervals — 74.5% during the day and over 80% at the closing batch. Importantly, we do not find evidence of intra-day momentum strategies among foreign investors since L1M is indistinguishable from zero. In other words, the trading intensity of foreign investors in the closing auction does not depend on the stock returns during the

32

For example, contagion trading might be more prevalent among countries with important trade links or similar market fundamentals. In Appendix B, we provide complementary evidence that flows among passively-managed foreign funds are strongly correlated in the cross-section. That is, managers of foreign index and passive funds trade more intensively on days when other passive funds are trading in the same direction, a behavior that cannot be explained by momentum strategies. 33

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

21

Table 13 Momentum trading by interval. L0M and L1M are calculated based on trades and returns during the continuous trading session and the closing batch auction separately. The L0M is the contemporaneous correlation between trades and stock returns during the interval. The L1M measure of momentum investing is based on the stock returns in the previous interval. ∗∗ ∗ /∗∗ /∗ indicate that the coefficient estimates are significantly different from zero at the 1%/5%/10% level. t-statistics in parentheses.

Complete day session

Continuous trading

Closing batch auction

Index

Passive

Active

Index

Passive

Active

Index

Passive

Active

Foreign L1M

Average

0.06∗ (1.78) 74.2

0.03∗ (1.69) 58.6

0.06∗∗ ∗ (3.68) 60.7

0.21∗ (1.67) 72.6

0.07∗∗ (2.22) 62.4

0.07∗∗ (2.47) 61.7

0.01 (1.27) 64.5

0.01 (0.72) 52.5

−0.00 (0.68) 50.5

L0M

Average

0.13∗∗ ∗ (3.48) 79

0.16∗∗ ∗ (5.17) 72

0.02 (1.00) 60

0.30∗∗ (1.98) 74.5

0.10∗∗ (2.42) 70.8

−0.03 (0.92) 54.6

0.03∗∗ ∗ (3.35) 80.60

0.03∗∗ (2.33) 69.10

−0.04∗∗ ∗

−0.02

−0.19∗ ∗ ∗

−0.18∗∗ ∗

−0.40∗ ∗∗

Percentage positive

(8.03) 34

(13.20) 33.9

−0.12 (1.45) 48.40

−0.18∗∗ ∗

(0.44) 47.7

(5.83) 34.9

(12.69) 28.7

0.00 (0.10) 41.90

−0.02 (1.44) 41.70

0.00 (0.25) 39.40

Average

−0.17∗∗ ∗

−0.22∗ ∗ ∗

−0.41∗∗ ∗

−0.30∗ ∗ ∗

−0.42∗∗ ∗

−0.66∗ ∗ ∗

−0.14∗∗ ∗

−0.16∗ ∗ ∗

−0.19∗∗ ∗

Percentage positive

(3.81) 30.8

(10.42) 33.5

(21.23) 26.9

(3.52) 21.9

(10.25) 24.2

(19.21) 26.4

(2.74) 19.4

(7.88) 20.8

(11.66) 21.5

Percentage positive

Percentage positive Domestic Institutions L1M Average

L0M

(3.58) 42.10

continuous trading session. Domestic institutional investors, on the other hand, trade in the opposite direction of returns in both intervals. Equity flows by passively-managed foreign funds seem to generate a significant price impact, an effect that is not driven by intra-day return chasing. Since a large portion of the price effect appears to be transitory, the evidence is consistent with liquidity frictions in the domestic market. Overall, funds that try to accommodate large flows in relative proportions to the benchmark index before the end of the trading session experience large transaction costs. 6. Conclusions In this paper, we examine the performance of investors separated by borders with different levels of active management. We find that the aggregate under-performance of foreign investors in the Colombian Stock Market is largely attributable to the behavior of foreign funds with passive management strategies (i.e. index funds and those with holdings similar to a domestic index). These funds pay higher prices for stocks purchases, receive less when they sell, and display inferior risk-adjusted returns relative to other investors. Market microstructure characteristics, such as average trade size, market impact, and bid-ask spreads, constrain the speed at which managers can execute their trades. If passive funds follow common signals and their flow imbalances are large, in order to trade stocks in similar index proportions, managers pay higher transaction costs to meet the end-of-day “deadline,” that is, they pay a price to liquidity providers to increase the speed of their trades to stay close to their benchmark. It is possible that the global strategies of passively-managed foreign funds are optimal, and the high transaction costs in one of their target countries, especially an opaque one, are offset by the gains in other markets. For example, if they follow cross-country hedging strategies. Alternatively, uninformed individuals seeking to invest in global markets might self-select into explicit index funds (perhaps attracted by their low fees), or uninformed managers of self-declared active funds might decide to track their respective benchmarks closely as an index fund would have. In such cases, large responses of less informed investors to news coming from international markets might be suboptimal if trading costs are generating a sizable drag in portfolio performance. Whether these foreign investors are behaving optimally, or whether they could benefit from trading more smoothly to reduce their transitory price impact remains an open question. Our single-country analysis, however, does highlight the costs of mechanical trading strategies in small markets. We anticipate that similar findings are likely in other developing countries, or in illiquid stocks in developed markets. In such cases, the costs of index investing might outweigh its benefits. Appendix C. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.finmar.2019.100509.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

22

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

Appendix A. Variable Definitions

Multimedia component 1 Variable

Definition

Investor/Fund information As defined by Cremers and Petajisto (2009). Calculated for each fund at the ActiveSharei,d beginning of the month. FundSizei,d Log of assets under managament at the beginning of the month. Minimum between yearly purchases and sells normalized by the average of FundTurnoverRatioi,d the fund’s assets. Trade value by investor i on stock s over total trade value of stock s on day d intensityi,s,d intervali,s,d Dummy variable equal to one for trades executed in the closing batch. NetFlowsi,d Daily fund net flows normalized by assets managed at the beginning of the month. Log number of stocks traded during the day in the same direction of s. stocksi,s,d Log number of trades during the day in stock s. tradesi,s,d Stock-Firm Characteristics Book-to-market ratio, computed as the total net assets, divided by the total B∕Ms,d market value of equity at the end of the quarter. Firm size computed as the log of total assets at the end of the quarter. lsizes,d ctc5s,d Close to close return of last five trading days. ctos,d Overnight returns between d − 1 and d. Open to close daily return. otcs,d PSs,d Price sensitivity, defined as ln(ph ∕pl )∕V olume where ph and pl are the highest and lowest price in each trading day.

Source CSE CSE CSE CSE CSE CSE CSE CSE

SFC, Datastream SFC, Datastream Bloomberg Bloomberg Bloomberg Datastream

Appendix B. Cross-sectional correlation in equity flows In this Appendix, we test whether investors’ demands by management style are correlated in the cross-section. To examine this question we use the dollar ratio measure of excess demand calculated for each investor following Lakonishok et al. (1992). For a given day d, Dratioi,d is defined as:

∑S Dratioi,d = ∑sS=1 s=1

buysi,s,d − sellsi,s,d buysi,s,d + sellsi,s,d

,

(B.1)

where buysi,s,d are total purchases on stock s by fund i, and sellsi,s,d are total daily sells. We begin the analysis by estimating, for each investor, a time-series regression of the investor daily excess demand (Dratioi,d ) on the excess demand by all investors in the same group of active management (Dratiod−i )34 Dratioi,d = 𝛼i + 𝛽i Dratiod−i + 𝜀i,d

for d = 1, ..., D

(B.2)

Estimation of equation (B.2) includes days in which fund i is active. The empirical strategy follows the same logic as the Fama-Macbeth methodology for each investor i. Here we run N time series regressions and calculate the mean and standard error of the 𝛽 coefficient for each group of active management, and separately for domestic and foreign investors. The average 𝛽 captures the extent to which daily flows are correlated across investors in the same group. We standardized both the dependent and independent variable such that both have zero mean and unit variance. The results are reported in the first column of Table B1. We find strong evidence that demands by index and passivelymanaged foreign funds are correlated in the cross-section. The coefficient associated with the contemporaneous demand of the foreign index funds is 0.11 and differs significantly from zero at the 1% level. Because the data are standardized, this suggests that a one standard deviation increase in the same day demand by other passive funds is associated with an increase in 0.11 standard deviations of passive fund flows. Since the regression only has one independent variable, the coefficient can also be directly interpreted as the cross-sectional correlation between daily equity demands among passively-managed foreign. For actively-managed foreign funds, the cross-sectional correlation between daily demands is positive but indistinguishable from zero. For domestic index funds, the daily correlation across demands is positive but smaller than for foreign index funds — close to one-third of the observed correlation for foreign index funds. It is possible that the daily cross-sectional correlation across demands by passive funds may result from momentum strate-

34

i To avoid spurious correlation between the two variables, we exclude fund i from the calculation of Dratio− . d

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

23

gies, if a large fraction of passive funds follow market returns.35 To evaluate the role of momentum trading in the relation between investor demands, we add lag standardized market returns to equation (B.2). In addition to the lag returns of the Colombian stock market we add standardized lag developed-market returns. This controls for the potential relation between market performance abroad and cross-border net flows at daily frequencies (Griffin et al., 2004). Average coefficients for each group of active management are reported in model (2) of Table B1. The main take away is that the cross-sectional correlation between contemporaneous demands changes little after accounting for domestic lag market returns and international lag returns. In other words, momentum trading does not appear to be the primary source behind the correlation across demands of passively-managed foreign funds. Table 1 Correlated demands by investor style. The table presents estimated coefficients of the linear regression i + 𝜀i,d (model 1), where Dratioi,d is excess demand of fund i on each day defined as Dratioi,d = 𝛼i + 𝛽i Dratio− d Dratioi,d =

∑S buysi,s,d −sellsi,s,d s=1 ∑S buysi,s,d +sellsi,s,d s=1

i and Dratio− is the excess demand by all investors in the same group of active management. d

Coefficients are averaged across investor time series regressions for each group of active management. ∗∗∗, ∗∗, ∗ indicate that the coefficient estimates are significantly different from zero at the 1%, 5%, 10% level. (1) Dratio Foreign Index Passive Active

Domestic Index Passive Active

(2) −i

Dratio

Numberof Funds −i

Lag Returns

Lag Developed Returns

0.11∗∗∗ (5.46) 0.09∗∗ ∗ (6.07) 0.02 (1.52)

0.11∗∗ ∗ (4.62) 0.09∗∗ ∗ (5.77) 0.02 (1.51)

0.03 (1.08) 0.04∗∗ (2.13) 0.01 (0.82)

0.03 (0.83) −0.04∗ (1.82) 0.01 (0.62)

0.04∗∗ ∗ (3.38) 0.01 (0.20) −0.00 (0.25)

0.05∗∗ ∗ (4.04) 0.01 (0.19) −0.01∗ (1.91)

−0.01 (0.38) −0.03∗ ∗ ∗ (3.50) −0.04∗ ∗ ∗ (6.3)

0.01 (0.62) −0.01 (1.34) −0.01∗ ∗ (2.42)

145 200 499

160 518 1425

References Agarwal, Sumit, Faircloth, Sheri, Liu, Chunlin, Rhee, S Ghon, 2009. Why do foreign investors underperform domestic investors in trading activities? Evidence from Indonesia. J. Financ. Mark. 12 (1), 32–53. Agudelo, Diego, Byder, James, Yepes, Paula, 2019. Performance and informed trading. Comparing foreigners, institutions and individuals in an emerging stock market. J. Int. Money Financ. 90, 187–203. Albuquerque, Rui, Bauer, Gregory H., Schneider, Martin, 2009. Global private information in international equity markets. J. Financ. Econ. 94 (1), 18–46. Avramov, Doron, Russ, Wermers, 2006. Investing in mutual funds when returns are predictable. J. Financ. Econ. 81 (2), 339–377. Barber, Brad M., Lee, Yi-Tsung, Liu, Yu-Jane, Odean, Terrance, 2009. Just how much do individual investors lose by trading? Rev. Financ. Stud. 22 (2), 609–632. Barberis, Nicholas, Shleifer, Andrei, Wurgler, Jeffrey, Comovement, , 2005. J. Financ. Econ. 75 (3), 284–317. Bollen, Nicolas PB., Busse, Jeffrey A., 2001. On the timing ability of mutual fund managers. J. Financ. 56 (3), 1075–1094. Brennan, Michael J., Cao, H Henry, 1997. International portfolio investment flows. J. Financ. 52 (5), 1851–1880. Calvo, Guillermo A., Mendoza, Enrique G., 2000. Rational contagion and the globalization of securities markets. J. Int. Econ. 51 (1), 79–113. Carhart, Mark M., 1997. On persistence in mutual fund performance. J. Financ. 52 (1), 57–82. Chan, Kalok, Covrig, Vicentiu, Ng, Lilian, 2005. What determines the domestic bias and foreign bias? Evidence from mutual fund equity allocations worldwide. J. Financ. 60 (3), 1495–1534. Chan, Louis KC., Lakonishok, Josef, 1995. The behavior of stock prices around institutional trades. J. Financ. 50 (4), 1147–1174. Chen, Honghui, Noronha, Gregory, Singal, Vijay, 2004. The price response to S&P 500 index additions and deletions: evidence of asymmetry and a new explanation. J. Financ. 59 (4), 1901–1930. Chen, Joseph, Harrison, Hong, Huang, Ming, Kubik, Jeffrey D., 2004. Does fund size erode mutual fund performance? The role of liquidity and organization. Am. Econ. Rev. 94 (5), 1276–1302. Choe, Hyuk, Kho, Bong-Chan, Rene, M Stulz, 1999. Do foreign investors destabilize stock markets? The Korean experience in 1997. J. Financ. Econ. 54 (2), 227–264. Choe, Hyuk, Kho, Bong-Chan, Ren, M Stulz, 2005. Do domestic investors have an edge? The trading experience of foreign investors in Korea. Rev. Financ. Stud. 18 (3), 795–829. Cremers, KJ Martijn, Petajisto, Antti, 2009. How active is your fund manager? A new measure that predicts performance. Rev. Financ. Stud. 22 (9), 3329–3365. Cremers, Martijn, Ferreira, Miguel A., Matos, Pedro, Starks, Laura, 2016. Indexing and active fund management: international evidence. J. Financ. Econ. 120 (3), 539–560. Dvok, Tom, 2005. Do domestic investors have an information advantage? Evidence from Indonesia. J. Financ. 60 (2), 817–839. Edelen, Roger, M., 1999. Investor flows and the assessed performance of open-end mutual funds. J. Financ. Econ. 53 (3), 439–466.

35 There is extensive empirical evidence suggesting that institutional investors are momentum traders (e.g., Grinblatt, 1995; Wermers, 2000; Sias, 2004). That is, institutional investors systematically sell stocks with negative returns and buy stocks with positive returns.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509

24

A. Pedraza et al. / Journal of Financial Markets xxx (xxxx) xxx

Fama, Eugene F., James, D MacBeth, 1973. Risk, return, and equilibrium: empirical tests. J. Political Econ. 81 (3), 607–636. Fama, Eugene F., French, Kenneth R., 1993. Common risk factors in the returns on stocks and bonds. J. Financ. Econ. 33 (1), 3–56. Froot, Kenneth A., Ramadorai, Tarun, 2008. Institutional portfolio flows and international investments. Rev. Financ. Stud. 21 (2), 937–971. Froot, Kenneth A., Oconnell, Paul GJ., Seasholes, Mark S., 2001. The portfolio flows of international investors. J. Financ. Econ. 59 (2), 151–193. Greene, Jason T., Hodges, Charles W., 2002. The dilution impact of daily fund flows on open-end mutual funds. J. Financ. Econ. 65 (1), 131–158. Greenwood, Robin, 2005. Short and long-term demand curves for stocks: theory and evidence on the dynamics of arbitrage. J. Financ. Econ. 75 (3), 607–649. Greenwood, Robin, 2008. Excess comovement of stock returns: evidence from cross-sectional variation in Nikkei 225 weights. Rev. Financ. Stud. 21 (3), 1153–1186. Griffin, John M., Nardari, Federico, Rene, M Stulz, 2004. Are daily cross-border equity flows pushed or pulled? Rev. Econ. Stat. 86 (3), 641–657. Grinblatt, Mark, Keloharju, Matti, 2000. The investment behavior and performance of various investor types: a study of Finland’s unique data set. J. Financ. Econ. 55 (1), 43–67. Grinblatt, Mark, Sheridan, Titman, Russ, Wermers, 1995. Momentum Investment Strategies, Portfolio Performance, and Herding: A Study of Mutual Fund Behavior. The American Economic Review, pp. 1088–1105. Hau, Harald, 2001. Location matters: an examination of trading profits. J. Financ. 56 (5), 1959–1983. Hau, Harald, 2011. Global versus local asset pricing: a new test of market integration. Rev. Financ. Stud. 24 (12), 3891–3940. Kaminsky, Graciela, Lyons, Richard K., Schmukler, Sergio L., Managers, , 2004. investors, and crises: mutual fund strategies in emerging markets. J. Int. Econ. 64 (1), 113–134. Karolyi, G Andrew, Ren, M Stulz, 2003. Are financial assets priced locally or globally? Handb. Econ. Financ. 1, 975–1020. Keim, Donald B., Madhavan, Ananth, 1997. Transactions costs and investment style: an inter-exchange analysis of institutional equity trades. J. Financ. Econ. 46 (3), 265–292. Kodres, Laura E., Pritsker, Matthew, 2002. A rational expectations model of financial contagion. J. Financ. 57 (2), 769–799. Lakonishok, Josef, Shleifer, Andrei, Vishny, Robert W., 1992. The impact of institutional trading on stock prices. J. Financ. Econ. 32 (1), 23–43. Leuz, Christian, Karl, V Lins, Warnock, Francis E., 2010. Do foreigners invest less in poorly governed firms? Rev. Financ. Stud. 23 (3), 3245–3285. Lewis, Karen K., 1999. Trying to explain home bias in equities and consumption. J. Econ. Lit. 37 (2), 571–608. McInish, Thomas H., Wood, Robert A., 1992. An analysis of intraday patterns in bid/ask spreads for NYSE stocks. J. Financ. 47 (2), 753–764. Newey, Whitney K., West, Kenneth D., 1987. A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55 (3), 703–708. Pstor, Lubo, Stambaugh, Robert F., 2002. Mutual fund performance and seemingly unrelated assets. J. Financ. Econ. 63 (3), 315–349. Pedraza, Alvaro, 2015. Strategic interactions and portfolio choice in money management: theory and evidence. J. Money Credit Bank. 47 (8), 1531–1569. Petersen, Mitchell A., 2009. Estimating standard errors in finance panel data sets: comparing approaches. Rev. Financ. Stud. 22 (1), 435–480. Portes, Richard, Helene Rey, 2005. The determinants of cross-border equity flows. J. Int. Econ. 65 (2), 269–296. Seasholes, Mark S., Zhu, Ning, 2010. Individual investors and local bias. J. Financ. 65 (5), 1987–2010. Sias, Richard W., 2004. Institutional herding. Rev. Financ. Stud. 17 (1), 165–206. Wermers, Russ, 2000. Mutual fund performance: an empirical decomposition into stock-picking talent, style, transactions costs, and expenses. J. Financ. 55 (4), 1655–1703. Williams, Tomas, Claudio, Raddatz, Sergio, Schmukler, 2017. International asset allocations and capital flows: the benchmark effect. J. Int. Econ. 108, 413–430. Yan, Xuemin Sterling, 2008. Liquidity, investment style, and the relation between fund size and fund performance. J. Financ. Quant. Anal. 43 (3), 741–767.

Please cite this article as: Pedraza, A., et al., Costly index investing in foreign markets, Journal of Financial Markets, https://doi.org/10.1016/j.finmar.2019.100509