INSIDERS, OUTSIDERS AND PERFORMANCE OF VIETNAMESE FIRMS45498

2 University of Economics and Business, Vietnam National University, Room 512 E4 Building, 144 Xuan Thuy, Cau Giay, Hanoi ABSTRACT The general consensus in the finance literature is tha

INSIDERS, OUTSIDERS AND PERFORMANCE OF VIETNAMESE FIRMS

Richard Beason 1 , Tu Tran Thi Thanh 2 , Dong Dao 2

1 Alberta School of Business, University of Alberta, Edmonton Alberta, T6G 2R6 Canada

2 University of Economics and Business, Vietnam National University, Room 512 E4 Building, 144 Xuan Thuy, Cau Giay, Hanoi

ABSTRACT

The general consensus in the finance literature is that a large proportion of inside ownership (defined as greater than 5% ownership shares by non-institutional holders, managerial holdings, founding family holdings, cross-shareholdings by affiliated firms and ownership by creditors) tends to be associated with poorer performance (as measured by ROE or ROA), when compared with firms with lower inside ownership, all else equal This need not be the case, however, if insiders act as monitors of the firm and have the same interest in returns as outsiders Using data from listed firms in Vietnam, we test the hypothesis that greater insider ownership has a negative impact on firm performance

1 INTRODUCTION

1.1 Motivation

Firms in Vietnam are characterized by a relatively large proportion of insider holdings (39% on average), defined as the proportion of shareholding greater than 5%, and heavy reliance on bank finance The heavy reliance on bank finance is typical among developing countries, due to under-developed corporate debt markets This combination of high insider ownership and reliance upon bank finance reduces the power of minority interests and results in a diluted takeover mechanism We expect this will affect corporate performance as measured by standard profitability variables, such as ROE, etc

This situation is similar to Japan during its high-growth period (1955-1990) Weinstein and Yafeh (1998) find that firms with close bank ties enjoy greater access to capital

* Corresponding author

Email address: rbeason@ualberta.ca

under a main bank system, but suffer in terms of performance They also find that such firms engage in less risky and profitable projects, in line with the findings of Nakatani (1984) Weinstein and Yafeh as well as Morck and Nakamura (1999) find that the benefits from bank centered relationships accrue to the banks rather than residual shareholders Mehrotra et al (2010) find a similar bank-centric effect in mergers and acquisitions in Japan

The literature on ‘entrenchment,’ however, suggests an alternative hypothesis with respect to investment behavior (Jensen, 1986, 2001) Whereas the findings of Weinstein and Yafeh suggest bank influence resulting in sub-optimal risk-taking

by borrower firms due to the inability of banks to diversify risk through their lending portfolio, the entrenchment literature suggests otherwise Instead, the ‘entrenched’ firm (that is closely held by insiders including banks) is insulated from takeover and tends to over-invest Certainly the high degree of investment by Japanese firms during the bubble period outside of core competency suggest that the latter problems of entrenchment outweigh those of risk aversion

Entrenchment, however, could lead to the opposite result if the insiders act as vigilant monitors For example, Korean corporate groupings or Chaebol are typically associated with a powerful founding family One could envision a situation where the founding family in such groupings performs the same function as an active takeover mechanism Working with data on accruals quality, Beason, Kim, Kim and Mehrotra (2018) can find no evidence for such a monitoring effect, but do find positive impact

on accruals quality in association with bank monitoring The likely explanation for this result is that in the absence of powerful independent shareholders, it is the banks who perform the monitoring function in Korea

In this paper, we test the relationship between inside shareholder ownership, outside ownership (foreign) and bank ownership on firm performance

1.2 Background of the Vietnamese economy

After reunification in 1975-81 the Vietnamese economy was unable to meet goals set out in its five year plan During the Economic plan of 1981-85 there were already shifts toward a mixed economy Economic data for the entire 1975-85 period are unreliable From the mid-1980’s the so-called ‘DoiMoi’ or economic opening reforms were initiated It is in this period that Soviet assistance to its partners began

to wane, due to budgetary constraints resulting from the Afghan war and the Soviet arms race with the US Vietnam became open to foreign investment and sought membership in international and regional organizations Today Vietnam is a member

of Asean and the WTO (joined 2007), and has various bilateral trade agreements During the initial years of liberalization, from 1987 until 1992, real economic growth was quite remarkable, but inflation was very high (see table below) After that inflation was brought under control, though the central bank (State Bank of Vietnam) is not independent Reliable data on government spending, revenue, deficit and debt was not available until 2000, but the level of debt/GDP is quite modest (see table below) Agricultural exports and tourism remain high growth industries, with manufacture for export by foreign electronics by foreign firms (Japanese, Korean, etc.) taking

on increasing importance Export orientation is quite important, with the level of

Exports to GDP reaching as high as 40% in some years after economic opening The corporate debt market remains thin, as with most developing countries, so bank finance is predominant As such, Vietnam can be considered a bank centered financial system, serving as part of the motivation for this paper It is difficult to estimate current levels of non-performing loans, but many believe that during the mini-crisis of 2011 the level might have been around 15% of outstanding loans, though officially the reported levels are much lower (Nguyen, Tu, 2017) Currently there is a real-estate and construction boom in Vietnam, and we believe the potential for another banking crisis exists Overall, however, obtaining reliable lending and loan performance data on the banking sector is difficult

Table 1 Summary statistics for the Vietnamese economy

Summary Statistics (IMF: ‘Report for selected countries and subjects’ 2018)

Year Real GDP Growth Inflation (%) Government Debt/GDP

1.3 Banks and corporate groups in Vietnam

The five largest banks in Vietnam, VietinBank, BIDV, Vietcombank, Agribank and Sacombank dominate many aspects of corporate finance in Vietnam VietinBank, BIDV and Vietcombank are typically thought of as state-owned, but VietinBank, for example, is 20% owned by Japan’s Tokyo-Mitsubishi UFJ bank group Vietcombank,

in turn, is 15% owned by Japan’s Mizuho Bank Sacombank is traded, though State interest is significant (Nguyen, Tu, 2017) In turn, these 5 largest banks hold shares

in other Vietnamese banks but do not themselves trade with the noted exception of Sacombank Of the 13 traded banks in our sample, through general practice, it is as though they exist in a Glass-Stegal world in that their equity holdings of other firms are below 1% of outstanding shares This result is not related to any regulatory policies Government policy, however, is in favor of all banks ultimately becoming joint-stock companies, so this situation is likely to change over time (Nguyen, Tu, 2017)

Large corporate groupings are also an important feature in Vietnam, much like Japan and Korea Like many developing countries, Vietnam can be considered a dual economy, in that it is characterized by both large domestic corporate groupings, often with foreign participation, large foreign manufacturing and other facilities, and also a very large number of small enterprises and an informal economy Among the large groupings FLC, Vingroup, FPT, Alphanam, Hoa Phat and REE are the most notable For our purposes these groups are important, in that they are often large shareholders (greater than 5%) of many listed firms Small enterprises typically exist outside the regulated credit market, but again, State policy is aimed at eliminating this duality

1.4 Model

As is standard in this literature, we assume that in equilibrium share prices reflect the present discounted value of the flow of expected earnings by the firm This can

be thought of as a standard dividend growth model of the typical form In addition to this, however, we allow for ownership structure and therefore control and governance

of the firm to impact the actual future earnings flows, thereby impacting returns Essentially, an empirical model would be of the form:

R = a+ B1(X1) + B2(X2) + E (1) Where:

R is a vector of normalized returns (ROE or ROA) for the cross section;

X1 is a vector of ownership measures;

X2 is a vector of control variables;

E is the error term

2 DATA, SUMMARY STATISTICS AND RESULTS

Our data are from the two major exchanges for 2018 A total of 729 firms actively traded during 2018, and we use data for those 729 firms for our analysis The key ownership and variables are described in Table2 below Unfortunately, the data provider only supplies data for the most recent year In future drafts we will hand

input in order to create a panel data set Table 2 shows a high degree of insider ownership Blocks (ownership holdings greater than 5% of shares) represent nearly 39% of shareholdings Concentration is a dummy variable that takes the value of

1 if there is one or more shareholders with greater than 10% ownership, and zero otherwise Its mean is 40% Foreign ownership averages 10.4%, which is meaningful, but is still swamped by insiders Those related to the Board of Directors (related person) holdings average nearly 1.5%, with CEO’s holding about 3.8% of shares of firms in the sample on average Despite the importance of bank finance in Vietnam, banks hold less than 1% of shares on average in our sample

Table 2.

N Range Minimum Maximum Mean DeviationStd Variance Statistic Statistic Statistic Statistic Statistic Std

Error Statistic Statistic

Own_Blockholding 729 1.354 0.000 1.354 0.3879 01117 0.3015 0.091

Own_Bank 729 0.7994 0.000 0.7994 0.0085 0.0026 0.0694 0.005

Own_Foreign 729 0.7758 0.000 0.7758 0.1037 0.0054 0.1446 0.021

Own_Concentration 729 1 0 1 0.40 0.018 0.490 0.240

Own Related Person 729 0.3577 0.0000 0.3577 0.0147 0.0016 0.0419 0.002

Own_CEO 729 0.6116 0.000 0.6116 0.0378 0.0029 0.0794 0.006

Valid N (listwise) 729

2.1 Null hypotheses and results

The simple null hypothesis would be that the ownership variables will have no impact

on firm performance, with the alternative hypothesis being that the various ownership variables will have a statistically significant impact on firm performance This does not reflect our priors, however As outlined in the introduction to the paper, we generally expect that greater ownership by insiders will have a negative impact on firm performance, consistent with the literature on entrenchment (Jensen, 1986, 2001) Specifically, our priors lead us to believe that block-holding, concentration, related persons and CEO ownership should be negatively related to firm performance, that foreign ownership will be positively related to firm performance, with bank ownership either positive or negative (positive if monitoring effect dominates, negative if banks are insiders) We also have some prior beliefs with respect to signs of some of the control variables, and these will be discussed below Results are presented in the following two tables

Some of the results in Table 3 are consistent with the entrenchment hypothesis, while others are more consistent with greater monitoring among insiders The same

is true of Table 4 with respect to ROE Generally speaking, analysis of entrenchment/ insider behavior focuses on ROA measures, due to potentially large movements in the value of shareholder equity This can be remedied by using panel data, which discuss below in terms of extensions The ROE results will be somewhat problematic for us given that Vietnamese equity prices rose by over 64% on average in 2017, and then fell by nearly 13% in 2018 While the ROE results would seem to be more

reliable given the higher regression F, both values are reasonably high given the number of observations We will therefore turn our initial attention to the similarities

in results for both the ROA and ROE dependent variables

Table 3 Regression results: ROA dependent variable as function of ownership and control variables, 2018 Calendar Year

Dependent Variable ROA Mean=0.105 Standard devia-tion=0.167 Independent variable Estimated coefficient p-value

Block holdings 0.036*** 0.003

Bank holdings -0.039 0.375

Foreign holding 0.063** 0.04

CEO holdings 0.106** 0.013

Concentration -0.011 0.135

Related holdings 0.192** 0.011

Debt to assets -0.003*** 0.000

Regression statistics Adj R-square=.077 Nobs=729 Regression F=7.46

* Significance at 10%;

** Significance at 5%;

*** Significance at 1%

In terms of similarities of results, block holdings, CEO ownership, ownership by those related to board members have a statistically significant positive impact on both measures of normalized profitability, while the normalized amount of debt is statistically negative in both cases We can reject the null hypothesis of no impact for regression results with respect to these variables for both dependent variable measures in these cases, so we will focus on these results first Unlike our priors, the role of insiders seems to be positive in the Vietnamese case for calendar year 2018 This would imply, in the absence of other evidence, that insiders in Vietnam serve

a monitoring role, exercising their relative power to ensure profitable functioning of the firm In terms of international comparison, this is in stark contrast to the post-war high growth period in Japan, where insiders functioned as in the entrenchment hypothesis or bank entrenchment hypothesis (Jensen, 1986, 2000; Morck and Nakamura, 1999; Beason, 1998)

The logic here is simple, though in contrast to the entrenchment hypothesis, and more in line with family related industrial groups as in Korea (Beason, Kim, Kim and Mehrotra, 2018) That is, while a high degree of entrenchment in terms of either block holdings or concentration might allow managers to deviate from profit maximization, alternatively, the close block/family holdings might serve as an effective monitoring mechanism, given that family wealth is a function of firm performance Thus, our results linking block holdings, CEO ownership, and related ownership to firm

performance with respect to both dependent variables would reflect this relationship between close ties to the firm and a strong monitoring relationship

Some other observations are useful Very high concentration, such as shareholders

or blocks holding a 10% or greater stake is statistically insignificant regardless of the profitability measure chosen In the case of our sample, this essentially reflects that all of the positive monitoring effects are already captured in the block holding measure In general, however, one should expect this relationship to be non-linear, with negative impact beyond some level of ownership Almost certainly, our inability

to find a statistically significant negative impact beyond some level of ownership is the result of the data set being a simple cross section

The dichotomy of results with respect to foreign (purely independent) share-holding

is almost certainly due to the impact of cross-sectional versus pooled analysis Almost certainly foreigners are the purest form of outsider, and the extent of their shareholding should have a positive impact on profitability We find this with respect

to the ROA measure but not ROE Again, this is likely related to the fact that our sample is a one-year cross section For example, 2017 was a year of extreme returns, with stock prices rising by nearly 65% as measured by MSCI On the other hand, after profit taking, 2018 was characterized by a 12.7% decline in share prices averaging over our two exchanges as measured by MSCI This impact of profit taking will have reduced our denominator measure for ROE, and therefore affected the measured impact of foreign shareholders for our sample year Again, this is a limitation of our simple cross section, and we expect that once we are able to extend to a panel data set there will be consistent results in terms of foreign ownership and profitability measures

Table 4 Regression results: ROE dependent variable as function of ownership and control variales, 2018 Calendar year

Dependent Variable ROE Mean=.055 Standard deviation=.083 Independent Variable Estimated Coefficient p-value

Constant -0.274*** 0.003

Block holdings 0.071*** 0.001

Bank holdings -0.109 0.180

Foreign holdings 0.025 0.546

CEO holdings 0.206*** 0.01

Concentration -0.01 0.473

Related holdings 0.459*** 0.001

Debt to assets -0.012*** 0.000

Firm size 0.03*** 0.000

Regression Statistics Adj R-sqiare=.202 Nobs=729 Regression F=23.991

In terms of control variables, we have no strong opinions about firm size, except

that firm size will have an impact on liquidity, thereby possibly affecting profitability Again, however, we find a positive and significant impact in terms of ROE (which should be affected by market interpretation of the likely positive impact of size-liquidity on profitability) whereas the impact on ROA is nugatory (quite reasonable) Debt structure, on the other hand, is fully consistent in terms of impact on both ROE and ROA (negative), as should be expected Certainly all else equal, debt charges (a cost variable) must negatively impact on any normalized profitability measure This control variable serves to ensure reasonable specification of our empirical models

2.2 Extensions

Clearly, while our results are reasonable, the apparent dichotomy in the results with respect to foreign shareholdings is troublesome In nearly every study of ownership structure, control and performance, foreign shareholdings are the most ‘outside,’ and therefore are associated with positive impact on profitability or returns We do not doubt that this relationship should extend to Vietnam Our results with respect

to the impact of foreign shareholding are almost entirely the result of a statistical artifact of foreign profit taking and retrenchment after the bumper year of 2017 Therefore, our major extension will be to hand collect/input data for the additional 4 year period for a total of 5 years and 729 firm observations

A further extension is comparability As outlined in the introduction, we have some basis for comparison with similar research with respect to Japan and Korea In terms

of comparison with Japan, our results are generally at odds, at least in terms of ownership structure, bank/financial sector ownership and firm performance In the case of Japan, insiders generally exert a negative influence (or at least until the 2000’s), including the role of banks In Korea, focusing on quality of accruals estimates (rather than general firm performance), it is found that insiders have a positive influence, perhaps as concerned monitors Our results in this paper are compatible with those results In future extensions, we will seek to pool data for Vietnam, China, Japan and Korea in order to understand similarities but also differences in terms of ownership structure, control and performance

3 CONCLUSIONS

In this paper we have attempted to place Vietnam within the general context of the ownership-control and entrenchment debate Our results are more generally consistent with insiders serving as monitors as opposed to insiders acting as rent seekers Our findings with respect to pure outsiders, such as foreigners, are sensitive

to the choice of dependent variable We believe this is due to our use of single year cross sectional analysis as opposed to panel data In future extensions to this paper

we plan to address this issue In sum, however, the Vietnamese stock market does not appear to be negatively affected by insider influence Indeed, insiders appear to act as positive monitors This is generally speaking, consistent with previous findings with respect to large Korean groupings, but inconsistent with Japanese groupings

[1] Aoki, Masahiko, 1990 Toward and economic theory of the Japanese firm Journal of

Economic Literature, 6: 27-43.

[2] Aoki, Masahiko, Hugh Patrick and Paul Sheard, 1994 The Japanese main bank system:

an introductory overview in Masahiko Aoki and Hugh Patrick, eds: The Japanese Main

Bank System, (Oxford).

[3] Beason, Dick (1998), ‘Keiretsu affiliation and share price volatility in Japan,’ Pacific

Basin Finance Journal, Vol 6, pp.27-43.

[4] Beason, Dick, Dong-Soon Kim, Renee Kim and Vikas Mehrotra (2018), ‘Accruals Quality and the role of insiders in Korea, working paper.

[5] Hoshi, Takeo, Anil Kashyup and David Scharfstein (1993), ‘The role of banks in

reducing the cost of financial distress in Japan,’ Journal of Financial Economics, Vol

27 pp 67-88.

[6] Jensen, M., 1986 Agency Costs of Free Cash Flow, Corporate Finance and Takeovers

American Economic Review Papers and Proceedings, 76: 323-329.

[7] Jensen, M., 2001 Value Maximization, Stakeholder Theory and the Corporate Objective

Function European Financial Management, 7: 297-317

[8] Kaul, Aditya, Vikas Mehrotra and Randall Morck, 2000 Demand Curves for Stocks

Do Slope down: New Evidence from an Index Weights Adjustment The Journal of

Finance, 55: 893-912

[9] Mehrotra, Vikas, Dimitri van Schaik, Jaap Spronk and Onno Steenbeek, 2010 Creditor-Focused Corporate Governance: Evidence from Mergers and Acquisitions in Japan

Journal of Financial and Quantitative Analysis, 2010

[10] Morck, R., and M Nakamura “Banks and Corporate Control in Japan.” Journal of

Finance, 54 (1999), 319-339.

[11] Nakatani, Iwao (1984), ‘The role of financial corporate groupings,’ in Masahiko Aoki,

ed: Economic analysis of the Japanese firm (North Holland, New York).

[12] Nguyen Thi Nhung, Tu Tran Thi ((2017), ‘Overview of Vietamese Banking,’ Conference Proceedings, Conference on Banking, Vietnam National University.

[13] PricewaterhouseCoopers (2017), “The World in 2050” (PDF).

[14] Shleifer, Andrei, 1985 Do Demand Curves for Stocks Slope Down? The Journal of

Finance, 41:579-590.

[15] Weinstein, David and Yishay Yafeh (1998) ‘On the costs of a bank centered financial

system: evidence from the changing main bank relations in Japan,’ Journal of Finance,

Vol LIII, No 2, pp 635-671.

[16] WorldBank.org (2017) “VDR.pdf” (PDF).