CC BY
17
ORIGINAL PAPER
DOI: 10.26794/2587-5671-2023-27-6-54-66 JEL G11, G14, G17, G24, L25
(CO ]
Effect of Underwriter's Reputation on Performance of Small Business IPOs
P. Siwacha, P. R. Kumarb, V. Guptac
a National Institute of Food Technology Entrepreneurship and Management, Kundli, India; a Amity University, Punjab, India; b National Institute of Food Technology Entrepreneurship and Management, Kundli, India; b University of Hyderabad, Hyderabad, India; c O. P. Jindal Global University, Sonipat, India
ABSTRACT
The purpose of the study is to determine the impact of the underwriter's reputation on shaping the short- and long-term IPO success of small businesses. The paper uses IPO data from 2012 to 2020, three reputable proxy and event-time methodologies to analyze the company's performance through market-adjusted excess returns, cumulative abnormal returns and buy & hold returns. Similarly, to mitigate common predispositions, use the calendar-time methodology, Fama-French three-factor model and Carhart four-factor model with high and low reputational groups. The study revealed a significant positive impact of underwriters' reputations on first-day returns and long-term performance. Overall, the results indicate that, in the long-term, IPOs of Indian small and medium-sized enterprises are over efficient with regard to reputation signals. On the contrary, the calendar-time method and multifactor model indicate the low long-term IPO effectiveness of the SME. According to the authors' conclusion, this is the first study to assess the impact of underwriter reputation on business performance using several reputation indicators, the calendar time methodology, and the multi-factor model on the ICP's Indian IPO platform. Keywords: SMEs; underwriter reputation; calendar time portfolio; underpricing; emerging market
For citation: Siwach P., Kumar P. R., Gupta V. Effect of underwriter's reputation on performance of small business IPOs. Finance: Theory and Practice. 2023;27(6):54-66. DOI: 10.26794/2587-5671-2023-27-6-54-66
ОРИГИНАЛЬНАЯ СТАТЬЯ
Влияние репутации андеррайтера
на результаты первого публичного размещения акций
компаний малого бизнеса
П. Сивача, П. Р. Кумар", В. Гуптас
a Национальный институт предпринимательства и управления пищевыми технологиями, Кундли, Сонипат, Индия;
a Университет Амити, Пенджаб, Индия; b Национальный институт предпринимательства и менеджмента в области пищевых технологий, Кундли, Индия; b Университет Хайдарабада, Хайдарабад, Индия; c Глобальный университет имени О. П. Джиндала, Сонипат, Индия
АННОТАЦИЯ
Цель исследования - определить влияние репутации андеррайтера на формирование краткосрочного и долгосрочного успеха IPO малых предприятий. Используются данные по IPO с 2012 по 2020 г., три репутационных прокси и методология событийного времени для анализа эффективности компании через скорректированную на рынок избыточную доходность, кумулятивную аномальную доходность и доходность от покупки и удержания. Аналогичным образом для смягчения общих предрасположенностей авторы используют методологию календарного времени, трехфакторную модель Фамы-Френча и четырехфакторную модель Кархарта с группами с высокой и низкой репутацией. Исследование показало значительное положительное влияние репутации андеррайтеров на доходность первого дня и долгосрочные показатели. В целом результаты свидетельствуют о том, что в долгосрочной перспективе IPO индийских малых и средних предприятий характеризуются избыточной эффективностью в отношении репутационных сигналов. Напротив, метод календарного времени и многофакторная модель свидетельствуют о низкой долгосрочной эффективности IPO МСП. Насколько известно авторам, это первое исследование, в котором изучается влияние репутации андеррайтеров на результаты деятельности компании с использованием нескольких репутационных показателей с применением методологии календарного времени и многофакторной модели на индийской площадке IPO МСП. Ключевые слова: МСП; репутация андеррайтера; портфель календарного времени; недооценка; развивающийся рынок
Для цитирования: Siwach P., Kumar П. P., Gupta V. Effect of underwriter's reputation on performance of small business IPOs. Финансы: теория и практика. 2023;27(6):54-66. DOI: 10.26794/2587-5671-2023-27-6-54-66
© Siwach P., Kumar P. R., Gupta V., 2023
BY 4.0
INTRODUCTION
Small business ventures are the backbone of any nation, as they contribute directly to the economy by providing employment, production and trade that further improve the people's standard of living and provide resources to the governments to run smoothly. Small business ventures continuously look for high-stake capital infusions to achieve their growth targets and expand their businesses. Therefore, the consistent fund requirement for small business ventures results in the establishment of alternative stock exchanges worldwide. On the contrary, these small ventures also suffer an identity crisis in the capital market. They face a lack of investor attention due to information asymmetry and high market uncertainty. Thus, underwriters contribute significantly by handling IPOs, signaling IPO quality, mitigating information asymmetry, and promoting IPOs to assure better performance.
In emerging economies, the capital market for small-medium enterprises (SMEs) suffers heavily due to a lack of investor attention, larger lot sizes, high volatility, ex-ante uncertainty and relaxed disclosure for SME IPOs. It further causes information asymmetry for investors and issuers. Small businesses thus leverage the underwriter's reputation to communicate the quality of the IPO to stakeholders. While literature on underwriter reputation has received wide academic attention, much less is known in contrast with the performance of SME IPOs in emerging markets. In international literature, researchers used event studies to capture IPO performance. However, using event time methodology (ETM) brings along new listing bias, resampling bias, skewness bias and potential issue of cross-sectional dependency of the returns during performance assessment. Hence, previous studies used the calendar-time portfolio approach to mitigate these biases and have robust results [1, 2].
In India A. Bhattacharya [3] examined the IPO performance of 106 SME IPOs from 2012 to 2015 and found that well-timed IPOs and underwriter reputation affect long-term success. However, study employed BSE small caps to determine the UDP of the sample IPOs. N. Arora and B. Singh [4] covered this gap by using BSE SME IPOs as benchmark proxy and found long-term IPO overperformance using cumulative abnormal returns (CAR) and buy & hold abnormal returns (BHAR). They recommended utilising the calendar-time portfolio methodology to calculate long-term performance in future research. Thus, the study contributes to the literature by using calendar-time portfolio and multifactor models to
examine the long-term performance of SME IPOs. Similarly, we calculated MAER, CAR, and BHAR from 2012 to 2020. We are using MAER and one-year BHAR to measure short-and long-run performance. Additionally, we are using multiple reputational proxies, such as registered capital (RPP1), market share based on IPO managed (RPP2) and market share based on capital raised (RPP3) to test the impact on long-run performance.
THE INSTITUTIONAL ENVIRONMENT OF THE INDIAN SME IPO PLATFORM
The Indian platforms provided a capital infusion to small businesses from the primary market. Multiple listing conditions and differences of these platforms are presented in Table 1.
DATA COLLECTION AND RESEARCH FRAMEWORK Model Description
The study sample involves 383 IPOs listed on the BSE SME and NSE Emerge platforms from 2012 to 2020. The data has been collected based on variables used in hypothesis formation. The data was collected from the Indian SME exchange, Ace Equity, and prospectuses.
Table 2 provides an overview of IPOs from 2012 to 2020 for their initial return (IR) and market-adjusted excess return (MAER). The IR is computed as the percentage difference between the initial day offer and the close price. Similarly, the MAER is calculated by the difference between the offer price and the closing price of the first day in percentage and then adjusted with the market return of the benchmark of the same day. The market benchmark for BSE and NSE started from 2013 and 2017, respectively.
Measuring IPOs performance
Event Time Method
The study uses ETM for evaluating the IPO's performance and computes post-issuance 1 to 36 months BHARs with the exclusion of IR. Following J. R. Ritter [5], month includes 21 consecutive trading days, with month-1 encompassing trading days 2-22, month-2 encompassing trading days 23-43, and so on. Therefore, the event year consists of 252 trading days and BHAR. t for IPO i is described as:
BHARi ,t =
nl (1 + Ri ,t )-'HnL (1 + Rm,t )- 1
(1)
In equation (1), R. t indicates daily returns of the stock i and R , denotes the market return of the
Table 1
Regulatory Framework of Indian SME IPO Platform
Particulars BSE SME NSE Emerge Main Board
Face Value (aftermarket paid-up capital) Less than INR 250 million. Less than INR 250 mil More than INR 100 mil.
Underwriting Should be underwritten 100%, with a minimum of 15% of IPO on the books of underwriters Should be underwritten 100%, with a minimum of 15% of IPO on the books of underwriters Mandatory (an exception to a condition where QIBs have been offered more than 50% of IPO)
IPO Gradation Not required Not required Required
Size of issue No restriction No restriction No restriction
Market making Required Required Not required
DRHP scrutiny Exchange Exchange Security Exchange Board of India
Reporting 6 monthly 6 monthly Quarterly
Track record 3 years 3 years Avg. operating profit of INR 150 mil. (pre-tax) each for any 3 years out of 5 years
Source: Indian SME Exchange.
Table 2
IPOs Distribution by Listing Year and Market-Adjusted Returns
Year S&P BSE SME IPO NSE SME Emerge Full market
N IR, % MAER, % N IR, % MAER, % N IR, % MAER, %
2012 1 2.8 0.00" 0 0.00" 0.00" 1 2.80 0.00"
2013 19 21.63 23.68 3 -0.12 8.91" 22 10.76 23.68
2014 17 11.45 13.09 2 -10.17 5.47" 19 0.64 13.09
2015 36 5.5 4.87 2 2.17 3.26" 38 3.84 4.87
2016 40 5.95 5.15 8 6.61 4.13" 48 6.28 5.15
2017 43 4.18 4.71 37 8.23 9.29 80 6.21 7.00
2018 62 3.63 3.47 92 7.33 7.5 154 5.48 5.49
2019 53 4.23 3.39 53 5.12 4.43 106 4.68 3.91
2020 19 16.53 16.3 6 2.39 2.34 25 9.46 9.32
Full sample 290 8.43 9.33 203 2.70 5.67 493"" 5.57 9.06
Source: Compiled by the author.
Note:' Indicates absence of SME benchmark index. Thus, for 15 firms the index return of BSE SME IPO is undertaken; "" Out of 493 IPOs, 110 issuing firms had either shifted to the mainboard or delisted.
benchmark for time t, respectively. Similarly, the average (BHARt) of the portfolio is as follows:
- 1 x-i N
BHAR, =—Y tBHARi t.
t n =1 ' t
(2)
The long-run returns can have a high potential for suffering from skewness bias. Therefore, to control this bias, J. D. Lyon et al. [6] suggested using bootstrapped skewness-adjusted test statistics, which can be computed as follows:
t (BHAR ) = 4N
S +1 yS2 3 '
1
6Nt
y
where S =
BHAR,t a (BHAR)
and y =
Y ¿(BHAR,t - BHARt) Nt a (BHARt )3
(3)
(4)
(5)
where y denotes the coefficient of skewness, Nt is the number of IPOs and a(BHAR) denotes the cross-section variation of yields in the sample [4].
THE CALENDAR TIME METHODOLOGY
We calculated the average abnormal monthly return for a diversified portfolio and further employed the E. F. Fama and K. R. French [7] three-factor and M. M. Carhart [8] four-factor models for evaluating the long-term returns and their time-series significance.
Rpt -Rf,, = a + ft Kt - Rf,t)+s ¡SMB,+hHML+e,,, (6)
In equation (6), R denotes monthly portfolio gains and R t indicates a risk-free rate for the month t, respectively. Similarly, Rm t is the market return of index and (Rm t - Rf t) signifies the market risk premium. Therefore, the multifactor model includes: stSMBt, which is the return difference of large and small-size portfolios, htHMLt denotes the difference in monthly portfolio returns concerning high and low B/M (book to market) portfolios.
RPt -Rf, =a+P;K,t-Rf,t)+
+ stSMBt + htHMLt + mtMOMt + e, ,t. (7)
In equation (7), miMOMl denotes the momentum factor, where momentum signifies difference in portfolio returns of winner and loser stocks. Similarly, p is the intercept to captures long-run performance. Thus, after controlling for (Rm t - Rf ,t), s tSMBt, htHMLt and mtMOMt, if a l is
different from zero, it indicates abnormal returns. Similarly, a positive and negative al can be interpreted as overperformance and underperformance in long-run.
Measuring Underwriter Reputation
The study uses three underwriter proxies to capture the different aspects of an underwriter's reputation. Following R. Michaely and W. H. Shaw [9], the study presumes that the scale of an investment bank can be associated with its reputation. Therefore, we use registered capital of the underwriters as RPP1. All 56 underwriters that managed at least one IPO were ranked into eight groups based on their registered capital. The underwriter with the highest registered capital is given a rank of 1, while the underwriter with the least registered capital is given a rank of 8. The RPP2 is based on the assumptions of C. Su and K. Bangassa [2]. They presume that underwriters who handle more IPOs are well-known in the market and have a better reputation. Thus, we assign one point for each IPO undertaken, and if multiple underwriters are involved, one point is distributed in equal proportion. Further, ranks are assigned from 1 to 8 based on issue counts, where rank 8 is given to underwriters with higher issue counts and rank 1 is given to underwriters with lesser IPO counts. Following W. L. Megginson and K. A. Weiss [10], we believe that higher capital raised by underwriters will result in a better market reputation. Thus, in RPP3, we use the ratio of each underwriter's gross proceeds to total gross proceeds.
Model Framework
We established MAER and 12-month BHAR as the dependent variables, whereas RPP1, RPP2, and RPP3 are used as independent variables with 8 control variables.
Issue size (SIZE), We computed the issue size based on the market offering of the IPOs. Prior literature shows a higher degree of ex-ante uncertainty for small-size IPOs compared to big-size issues due to the higher opacity of information [5, 11]. Therefore, we expect a positive association between an issue's size and the IPO's performance in the long term.
Initial returns (IR): Initial returns are unadjusted first-day raw returns of the IPOs. Prior literature stated the need to separate the first-day benchmark adjusted excess return and long-term performance from the initial returns, according to S. Song et al. [13]. S. Dhamija and R. K. Arora [14] find a positive association between IR and UDP and vice versa. Thus, we are assuming a positive association between IR and IPO performance in the long-run.
Firm Size (SIZE : Small firms publish little operating history, have low sales, high volatility, and low reputational underwriters, which creates higher information asymmetry and speculation and causes longrun underperformance [3, 5, 15]. So, we presume a positive association between firm size and IPO performance.
Total Subscription (T_SUB): The total subscription is computed as the subscription rate of the IPO by the investors. According to fad theory, over-optimism may also cause investor overreaction and speculation that can lead to lower long-run performance [4, 16]. Thus, the study assumes a negative relationship among total subscriptions and IPO performance in the long term.
Earnings per share (EPS): F. Allen and G. R. Faulhaber [17], through signaling theory, state that good quality companies differentiate themselves from bad quality firms through underpricing the IPO. Hence, the study includes EPS to measure firm quality, operating performance and expects a positive relationship with long-run performance.
Debt ratio (DA): The debt ratio indicates the issuing company's leverage and the firm's overall financial risk [12]. Therefore, a higher ratio causes a higher risk, resulting in long-run underperformance and vice versa. We expect a negative association between debt ratio and IPO performance.
EBITDA to asset ratio (EBITDAA): The EBITDA to asset ratio is the profitability measure of the firm. This ratio includes the overall income of the firm in place of operating income, thus providing a clearer picture of the earnings [12]. The higher EBITDAa ratio indicates a firm's better profitability position, thus resulting in better long-run performance. Therefore, we assume a positive relationship between the EBITDAa ratio to long-run performance.
Market Condition (MKTCD): Following N. Arora and B. Singh [4] and P. Jaskiewicz et al. [18], we explained the hot market by creating a dummy variable. The managers take advantage of favourable market conditions and try to time their IPOs accordingly, creating investor overvaluation that further leads to higher first-day returns but lower returns in the long term. Therefore, we expect a negative association between the hot market and IPO performance.
Table 3 presents descriptive statistics and a correlation matrix for selected variables. Similarly, analysing the cross-sectional association of MAERs and one-year BHARs with the selected variables is described in the following equation:
MAER = a + P1 [all reputational proxies ]+P2SIZE/ + + P3IR + p 4SIZEf + P5TSUB + p6 DA+P7EBITDAA + (8)
+ P8eps+p9mkt+e,
BHARl2m = a + P: [all reputational proxies] + P2SIZEi + + P3IR + P4SIZEf + P5TSUB + P6DA + P7EBITDAA + (9) + P8EPS + p9 MKT + e.
Hypothesis 1:
Hj: The reputation of underwriters has a positive impact on MAERs of the Indian SME IPOs.
Hypothesis 2:
Hj: The reputation of underwriters has a positive impact on one-year BHARs of the Indian SME IPOs.
DATA INTERPRETATION AND DISCUSSIONS
Analyzing Long-Run Performance Using Event Time Methodology
The ETM uses CARs and BHARs to evaluate the IPO's performance in the long term. Table 4 presents the values of CAR and BHAR for event time t in columns 3 and 5, respectively, from 1 to 36 months. Although a negative value indicates IPO underperformance, a slight improvement can be seen from the 1st month to the 24th month. Similarly, MAERs shows -0.25% return for first month, which further increases by 61.37%, to 209% from 3 to 36 months.
In Table 5, we classify MAERs and 12m BHARs with respect to RPP1, RPP2, RPP3 and divide them into groups of high and low reputations. Panel A exhibits that low-reputation underwriters had a more significant rate of UDP than reputed underwriters. The result can be explained by a lack of quality signals among less reputed underwriters and concern for lower subscription levels [19]. Additionally, the average one-year BHARs are comparatively higher for less reputed underwriters. A similar trend is exhibited in panel B, where high and low reputational measure is based on issues managed by underwriters. Hence, underwriters managing more than 4% of the IPOs are categorized as having a high reputation and lower reputation otherwise. In panel C, underwriters with more than 7% of total proceeds indicate a strong reputation. The MAERs for RPP3 is 6%, identical to the mean of whole sample, and the same is reflected in case of one-year BHARs.
Results of Cross-Sectional Regression
Table 6 exhibits the results of multivariate cross-sectional OLS regression based on equations 8 and 9. In panel A, the empirical results indicate that RPP1 and RPP3 statistically impact MAER. However, RPP2 is found to be inconsequential in row 2. The results are coherent when we are regressing the same reputational proxy for the entire dataset
Table 3
Correlation Matrix and Descriptive Statistics
rpp1 rpp2 rpp3 size. f ir sizej, t_sub da ebitdaa eps mktcd
Panel A: Descriptive statistics
N 383 383 383 383 383 383 383 383 383 383 383
Mean 4.5 3.32 7.43 12.18 -0.072 42.73 9.76 0.25 0.07 43.03 0.67
Min. 1 0 0.04 1.17 -28.5 0.6 0.07 0 -4.23 -530.56 0
Max. 8 7 19.8 92.74 30.28 696.13 263.09 2.97 4.42 7779.54 1
Std. Dev. 2.09 1.85 6.51 11.86 5.82 75.57 29.69 0.25 0.34 417.6 0.47
Panel B: Correlation matrix
RPP1 1
RPP2 0.151'" 1
RPP3 0.220"' .218"' 1
SIZEi -0.008 .142"' 0.094 1
IR -0.004 -0.061 0.009 -0.007 1
SIZE, -0.173"" 0.054 0.010 .354"" 0.010 1
T_SUB 0.127" -0.047 0.004 0.025 -0.093 -0.065 1
DA 0.008 -0.032 .128" 0.010 -0.079 -0.041 0.023 1
EBITDA_A 0.105" -0.004 -0.038 0.061 0.048 0.071 -0.003 -.384'" 1
EPS -0.020 0.023 .124" -0.025 -0.016 -0.029 0.012 -0.070 0.015 1
MKTCD 0.099 0.093 0.083 .216"' -0.036 -.102" .193"' -0.025 .135'" 0.040 1
Source: Compiled by the authors.
Note:"10% significant level, "5% significant level, """1% significant level.
Q c
■Q
Table 4
The Post-Issue CARs and BHARs for the Event Months
Event time (t) Nt CARt t (CARt) BHARt t (BHARt)
1 month 383 -0.0731 1.794 -0.0025 17.169""
3 months 383 -0.0571 2.526'"" 0.6137 8.691""
6 months 324 -0.0424 3.214""" 0.9503 7.761""
9 months 281 -0.0358 1.734 1.2571 4.922""
12 months 240 -0.0342 3.091"" 1.3787 5.203""
15 months 191 -0.0305 2.739"" 1.6467 5.545""
24 months 107 -0.0295 4.297"" 1.9002 1.611
36 months 29 -0.0436 3.005"" 2.0951 5.002""
Source: Compiled by the authors.
Note: "10% significant level, ""5% significant level, """1% significant level.
Table 5
The Measure of Reputational Estimates to Market-Adjusted Excess Return and One-Year BHARs
Reputational Proxies Benchmark adjusted returns
MAER 12-months BHARs
N X Median a N x Median a
Panel A: Reputational estimate of RPP1
High (RPP1 > 60 mil INR) 219 5.222 3.120 10.950 146 1.301 1.151 1.280
Low (RPP1 < 60 mil INR) 164 7.091 1.620 23.790 94 1.500 1.360 1.118
Panel B: a reputational estimate of RPP2
High (RPP2 > 4%) 266 5.949 3.355 12.349 166 1.315 1.166 1.281
Low (RPP2 < 4%) 117 6.190 1.100 25.958 74 1.522 1.522 1.068
Panel C: a reputational estimate of RPP3
High (RPP3 > 7%) 183 6.052 3.740 9.769 129 1.426 1.356 1.256
Low (RPP3 < 7%) 200 5.996 1.650 22.576 111 1.324 1.224 1.180
Full Sample 383 6.023 2.200 17.657 240 1.379 1.238 1.223
Source: Compiled by the authors.
in row 4. The results are consistent with those of N. Arora and B. Singh [20] and J. Gao et al. [21]. The argument behind the positive relationship can be the behaviour of reputed underwriters to manage the IPOs of low-risk firms to avoid negative listing and lack of subscription by the investors. Furthermore, the higher investor demand can be reflected in the elevated closing price as well as higher UDP levels [22]. Similarly, the reputed
underwriters used UDP to avoid insolvency risk and signal IPO quality [23].
Panel B of Table 7 exhibits the regression results of reputational proxies on one-year BHARs, where the individual regression outcome of reputational measures is shown in rows (5), (6), (7), the whole sample is shown in row (8). The result shows a significant positive impact of RPP1 and RPP2 on one-year BHARs. Similarly, RPP3 shows a positive but
Table 6
The Outcome of Cross-Sectional OLS Regression
Intercept RPP1 RPP2 RPP3 SIZEi IR SIZEf T_SUB DA ebitdaa EPS MKT Adj. R2
Panel A: MAERs used as a dependent variable for Indian SME IPOs
(1) 0.945 0.148 -0.165 0.654 0.217 0.464 0.169 -0.050 0.178 0.037 0.614
(1.98) "" (2.67) """ (15.05) """ (3.15) """ (10.49) """ (2.56) """ (0.73) (4.23) """ (0.750)
(2) 2.010 0.072 -0.159 0.652 0.203 0.473 0.168 -0.041 0.176 0.032 0.609
(0.94) (2.55) """ (14.92) """ (2.92) """ (10.74) """ (2.53) """ (0.61) (4.14) """ (0.65)
(3) 0.973 0.176 -0.159 0.636 0.187 0.487 0.146 -0.053 0.151 0.512 0.629
(3.26) """ (2.63) """ (14.83) """ (2.79) """ (11.29) """ (2.25) """ (0.81) (3.58) """ (0.61)
(4) 2.937 0.358 0.160 0.240 -0.080 0.735 0.02 0.106 3.340 -1.005 0.147 0.452 0.641
(2.61) """ (0.97) (4.11) """ (2.92) """ (14.96) """ (3.21) """ (11.09) """ (2.21) """ (1.02) (3.55) """ (0.68)
Panel B: 12 months Buy & Hold Abnormal Return used as a dependent variable for the sample of Indian SME IPOs
(5) 2.442 0.186 -0.291 0.182 -0.005 -0.223 0.372 0.238 0.043 -0.154 0.332
(2.94) """ (3.28) """ (2.98) """ (-0.05) (3.62) """ (4.22) """ (2.48) """ (0.745) (1.96) """
(6) 2.955 0.246 -0.314 0.141 0.058 -0.261 0.358 0.239 0.066 -0.162 0.352
(3.78) """ (3.59) """ (2.34) """ (0.654) (4.31) """ (4.12) """ (2.54) """ (1.14) (2.09) """
(7) 2.128 0.092 -0.308 0.173 0.045 -0.235 0.377 0.217 0.043 -0.134 0.306
(1.24) (3.40) """ (2.78) """ (0.49) (3.75) """ (4.19) """ (2.22) """ (0.71) (1.16)
(8) 3.475 0.08 0.29 0.14 -0.31 0.149 0.01 -0.241 0.33 0.247 0.15 -0.16 0.389
(2.27) """ (4.33) """ (1.81) (3.63) """ (2.53) """ -0.14 (4.08) """ (3.96) """ (2.70) """ (3.55) """ (-2.07)
Source: Compiled by the authors.
Note: "10% significant level, ""5% significant level, """1% significant level.
insignificant impact on long-run performance. The results align with the studies of H. Anderson et al. [24], N. Arora and B. Singh [25], S. Dhamija and R. K. Arora [26] and C. Su and K. Bangassa [2]. It indicates that the underwriters use their expertise to undertake quality IPOs and avoid riskier issues.
The positive IPO performance can also be attributed to reduced information heterogeneity among investors due to efficient IPO estimation and signalling the accurate market value of the firm by the underwriters [25]. Conclusively, the regression results accept hypothesis 1 (H1) that the underwriter's reputation positively impacts the MAERs of the Indian SME IPOs. Furthermore, the results also accept hypothesis 2 (Hj) that the underwriter's reputation positively impacts the one-year BHARs of the Indian SME IPOs.
Estimating the IPO Performance in the Long-Term Using Calendar-Time Methodology
We are using the calendar-time portfolio to overcome the biases that emerge from event-time returns. Therefore, a diverse portfolio is being created for every calendar month from 2012 to 2020. However, companies that issued IPOs in the last 36 months are included in the final portfolio. And tested with FF3 and CF4 models. Concerning to calendar-time portfolio, our null hypothesis assumes that the mean monthly abnormal returns are equal to zero. Therefore, a subsequent multifactor regression intercept is employed to test our null hypothesis. Panels A and B of Table 7 show the calendar time regression results using the FF3 model and CF4 model, respectively.
Consequently, RPP1, RPP2, RPP3 are categorized into high and low reputational groups. The result
Table 7
Regression Results Using the Calendar-Time Approach
Full Sample RPP1 RPP2 RPP3
High Low High Low High Low
Panel A: Results of the Fama-French three-factor Model
Intercept -0.115 -0.114 -0.116 -0.113 -0.118 -0.088 -0.115
t-stat (54.74) •" (43.01) ••• (34.06) ••• (44.06) ••• (33.11) ••• (35.50) ••• (36.83) •••
Rm-Rf 0.135 0.140 0.129 10.268 0.158 0.130 7.837
t-stat (13.05) ••• (10.32) ••• (8.05) ••• (0.13) (8.31) ••• (9.95) ••• (0.14)
SMB 0.138 0.136 0.142 11.797 0.126 0.069 11.202
t-stat (13.57) ••• (10.17) ••• (8.99) ••• (0.31) (6.72) ••• (5.35) ••• (0.11)
HML -0.030 -0.019 -0.045 -2.171 -0.046 0.051 -3.055
t-stat (3.17) ••• (1.54) (3.04) ••• (0.03) (2.58) ••• (4.15) ••• (0.02)
Adj.R2 0.050 0.052 0.047 0.049 0.052 0.037 0.052
Panel B: Results of Carhart four-factor Model
Intercept -0.114 -0.113 -0.115 -0.112 -0.117 -0.087 -0.113
t-stat (53.95) ••• (42.27) ••• (33.67) ••• (43.40) ••• (32.66) ••• (35.27) ••• (36.24) •••
Rm-Rf 0.154 0.163 0.143 0.146 0.176 0.155 0.133
t-stat (14.21) ••• (11.45) ••• (8.53) ••• (11.29) ••• (8.84) ••• (11.22) ••• (8.59) •••
SMB 0.160 0.162 0.157 0.165 0.146 0.091 0.184
t-stat (14.76) ••• (11.39) ••• (9.42) ••• (12.80) ••• (7.35) ••• (6.75) ••• (11.80) •••
HML 0.005 0.023 -0.019 0.012 -0.011 0.093 -0.009
t-stat (0.48) (1.57) (1.05) (0.86) (0.53) (6.48) ••• (0.55)
MOM 0.078 0.093 0.058 0.079 0.074 0.092 0.071
t-stat (5.81) ••• (5.27) ••• (2.81) ••• (4.97) ••• (3.01) ••• (5.56) ••• (3.71) •••
Adj.R2 0.053 0.056 0.048 0.052 0.055 0.041 0.054
Source: Compiled by the authors.
Note: "10% significant level, "5% significant level, """1% significant level.
shows an abnormally significant long-term IPO performance. Although the positive abnormal returns using ETM imply overperformance, the negative intercept from Table 7 indicates the long-run underperformance of the firms. A similar trend of significant negative intercept can be observed in underwriters' reputational measures that are further categorized into high and low reputational groups. The results indicates that the intercept significantly differs from zero in panels A and B.
The result shows the significant abnormal performance of the IPOs in the long-run. However, the exhibited long-run overperformance based on ETM is converted into underperformance while using the multifactor model. The negative intercept indicate that Indian SME IPOs are performing similarly to international evidence [2, 27, 28].
The model suggests a strong impact of the underwriter's reputation on the long-run performance of the firms. However, the small adjusted R 2 indicates the minimal data variance explained by the model.
Table 8
Endogeneity Testing Using Two-Stage Least Square Regression
Dependent Intercept Instrument RPP1 SIZE 1 IR SIZEf T_SUB DA ebitdaa EPS MKT R 2
First-stage regression:
RPP1 11.177 1.746 -0.165 0.019 -0.008 0.014 0.112 0.313 0.011 0.644 0.507
(4.81)- (1.09) (0.47) (2.57)... (2.09)... (0.13) -0.73 0.33 1.03
Second-stage regression:
MAER 14.459 -0.351 -0.914 -0.318 0.002 0.104 -1.857 5.761 0.002 -2.92 0.151
(0.56) (3.28)" (2.98)-.. 0.15 (3.44)... (0.48) (1.96) ... 0.69 (1.41)
BHAR„ 12m 0.073 0.073 0.067 0.058 0.061 0.474 0.175 0.066 -0.381 0.605
(2.97)" 1.24 0.41 0.31 (2.81)... 1.49 (2.54)... 0.32 (2.90)...
Source: Compiled by the authors.
Note:" with parenthesis denotes white heteroscedastic-consistent t-stat, " 10% significant level, "" 5% significant level, """ 1% significant level.
Therefore, the fluctuations in performance using multifactor models and ETM support the arguments of P. A. Gompers and J. Lerner [29] and J. R. Ritter and I. Welch [30] that these deviations can be observed due to the sample selection criteria, time selection and methodology adopted to evaluate the IPO performance in the long run. We can infer that if investors are using ETM to assess returns, they will expect abnormal returns in long-term. However, in the case of a calendar-time portfolio, investors cannot expect similar abnormal returns in the long term.
Robustness Test
Multiple studies have suggested that the reputation of an underwriter can be an endogenous variable and therefore needs to be checked against the problem of endogeneity [2, 25, 31, 32]. Thus, we are using a two-stage least squares regression (2SLS) to check the robustness of our cross-sectional OLS regression and validate the results. Hence, we are constructing a new instrument variable based on the geographic difference between the issuer firm and the underwriting firm. We presume that due to the lower information cost, the underwriters with the shortest distance to the issuing firm have a higher probability of getting the opportunity to handle an IPO.
We used the address information of all the underwriters and issuing firms to create the instrument variable and confirmed the distance using Google Maps of India. In the case of an IPO underwritten by two or more underwriters, we use the distance of the leading underwriter to the issuing firm. At the first stage of the 2SLS model, we regress the instrument variable with the RPP1 using the same
control variable that we used in the Table 6 using the following equation:
RPP1 = a+P1 [INSTRUMENT ]+P2SIZEi + + P3IR + fi4SIZEf + PTTSUB + p6 DA + (10) + $7EBITDAa + P8EPS + P9MKT + £.
In the second stage of the 2SLS model, we generated the variable RPP1 from the regression coefficient of the first stage. Furthermore, we are using the predicted value of RPP1 instead of the value of reputational proxies used in sections (1) and (5) in Table 6.
MAER/BHAR = a+p1 RPP1+P2SIZE t +P3IR + PASIZEf + + P5TSUB + P6DA + P7EBITDAA +
+ P8EPS + P9MKT + £. (11)
Table 8 presents the two-stage least square regression to check for endogeneity issues and avoid issues with biased estimators in our regression results. The first stage regression of our instrument variable shows a significantly positive impact on UDP. The higher value of Wald f-statistics from its critical values shows the strength of instrument variable to explain the endogenous variable. Similarly, the explanatory power of these variables can be reflected through the McFadden R 2 of 50.70. The result supports positive impact of underwriter reputation on the IPO performance in long-run. This also suggests that Indian underwriters benefit the issuing firms and investors by providing better long-run performance.
CONCLUSION
Most studies on Indian SME IPO platforms majorly used ETM with a single reputational measure (market share) to draw conclusions [4, 25, 33, 34]. Since the reputation of an underwriter can be based upon various indicators, we have taken three different reputational proxies to test the impact of the underwriter's reputation on the first day and long-run performance of the firms. The results suggest that due to the lack of investor attention and higher cost of investor entry, the reputed underwriters play a substantial role in signalling IPO quality and shaping better long-run performance.
Furthermore, the calendar time regression exhibits a statistically significant negative intercept that indicates a long-run underperformance of SME IPOs. However, the lower coefficient value of intercept and adjusted R 2 of the multifactor model
shows an inadequate power to explain the impact of underwriter reputation on long-run performance. These conflicting results are aligned with the previous studies of P. A. Gompers and J. Lerner [29], J. R. Ritter and I. Welch [30] and C. Su and K. Bangassa [2] that the evaluation of long-term IPO performance immensely hinges on the selection criteria, sample period selection and methodology adopted by the investors. Similarly, in the absence of a standardized long-run performance measure, the previous literature also directs inconsistent results and differs based on firms and regulatory environments. Thus, we conclude that the argument that an underwriter's reputation has a significant positive impact on MAER's longrun performance holds true. Since there is no single accepted method to measure long-run performance, future studies may explore this considerable gap to come up with unique long-run measures to evaluate the firm's performance.
REFERENCES
1. Barber B. M., Lyon J. D. Detecting long run abnormal performance. Journal of Financial Economics. 1997;43(3):341-372. DOI: 10.1016/S 0304-405X(96)00890-2
2. Su C., Bangassa K. The impact of underwriter reputation on initial returns and long-run performance of Chinese IPOs. Journal of International Financial Markets, Institutions and Money. 2011;21(5):760-791. DOI: 10.1016/j.intfin.2011.06.002
3. Bhattacharya A. Innovations in new venture financing: Evidence from Indian SME IPOs. Global Finance Journal. 2017;34:72-88. DOI: 10.1016/j.gfj.2017.02.001
4. Arora N., Singh B. The long-run performance of SME IPOs in India: Empirical evidence from Indian stock market. Journal of Asia Business Studies. 2020;15(1):88-109. DOI: 10.1108/JABS-10-2019-0305
5. Ritter J. R. The long-run performance of initial public offerings. The Journal of Finance. 1991;46(1):3-27. DOI: 10.1111/j.15406261.1991.tb03743.x
6. Lyon J. D., Barber B. M., Tsai C.-L. Improved methods for tests of long-run abnormal stock returns. The Journal of Finance. 1999;54(1):165-201. DOI: 10.1111/0022-1082.00101
7. Fama E. F., French K. R. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics. 1993;33(1):3-56. DOI: 10.1016/0304-405X(93)90023-5
8. Carhart M. M. On persistence in mutual fund performance. The Journal of Finance. 1997;52(1):57-82. DOI: 10.1111/j.1540-6261.1997.tb03808.x
9. Michaely R., Shaw W. H. The pricing of initial public offerings: Tests of adverse-selection and signaling theories. The Review of Financial Studies. 1994;7(2):279-319. DOI: 10.1093/rfs/7.2.279
10. Megginson W. L., Weiss K. A. Venture capitalist certification in initial public offerings. The Journal of Finance. 1991;46(3):879-903. DOI: 10.1111/j.1540-6261.1991.tb03770.x
11. Beatty R. P., Ritter J. R. Investment banking, reputation, and the underpricing of initial public offerings. Journal of Financial Economics. 1986;15(1-2):213-232. DOI: 10.1016/0304-405X(86)90055-3
12. Wadhwa K., Neupane S., Syamala S. R. Do group-affiliated firms time their equity offerings? Pacific-Basin Finance Journal. 2019;54:73-92. DOI: 10.1016/j.pacfin.2019.02.004
13. Song S., Tan J. S., Yi Y. IPO initial returns in China: Underpricing or overvaluation? China Journal of Accounting Research. 2014;7(1):31-49. DOI: 10.1016/j.cjar.2013.12.001
14. Dhamija S., Arora R. K. Initial and after-market performance of SME IPOs in India. Global Business Review. 2017;18(6):1536-1551. DOI: 10.1177/0972150917713081
15. Miller E. Long run underperformance of initial public offerings: An explanation. Department of Economics and Finance Working Papers, 1991-2006. 2000;(16). URL: https://scholarworks.uno.edu/cgi/viewcontent. cgi?article=1015&context=econ_wp
16. Aggarwal R., Rivoli P. Fads in the initial public offering market? Financial Management. 1990;19(4):45-57. DOI: 10.2307/3665609
17. Allen F., Faulhaber G. R. Signaling by underpricing in the IPO market. Journal of Financial Economics. 1989;23(2):303-323. DOI: 10.1016/0304-405X(89)90060-3
18. Jaskiewicz P., González V. M., Menéndez S., Schiereck D. Long-run IPO performance analysis of German and Spanish family-owned businesses. Family Business Review. 2005;18(3):179-202. DOI: 10.1111/j.1741-6248.2005.00041.
19. Álvarez-Otero S., López-Iturriaga F. J. Does corporate governance affect the valuation of Spanish IPOs? The role of ownership structure and the board of directors. Revista Espanola de Financiacion y Contabilidad = Spanish Journal of Finance and Accounting. 2018;47(2):214-241. DOI: 10.1080/02102412.2018.1429556
20. Arora N., Singh B. Impact of auditor and underwriter reputation on underpricing of SME IPOs in India. Management and Labour Studies. 2019;44(2):193-208. DOI: 10.1177/0258042X19829285
21. Gao J., Cong L. M., Evans J. Earnings management, IPO underpricing, and post-issue stock performance of Chinese SMEs. The Chinese Economy. 2015;48(5):351-371. DOI: 10.1080/10971475.2015.1067085
22. Sundarasen S. D., Khan A., Rajangam N. Signalling roles of prestigious auditors and underwriters in an emerging IPO market. Global Business Review. 2018;19(1):69-84. DOI: 10.1177/0972150917713367
23. Liu X., Ritter J. R. Local underwriter oligopolies and IPO underpricing. Journal of Financial Economics. 2011;102(3):579-601. DOI: 10.1016/j.jfineco.2011.01.009
24. Anderson H., Chi J., Wang O. S. IPO performance on China's Newest Stock Market (ChiNext). The Chinese Economy. 2015;48(2):87-113. DOI: 10.1080/10971475.2015.99321524
25. Arora N., Singh B. Do prestigious underwriters shape the performance of SME IPOs in India? Global Business Review. 2020. DOI: 10.1177/0972150920963216
26. Dhamija S., Arora R. K. Determinants of long-run performance of initial public offerings: Evidence from India. Vision. 2017;21(1):35-45. DOI: 10.1177/0972262916681243
27. Thomadakis S., Nounis C., Gounopoulos D. Long-term performance of Greek IPOs. European Financial Management. 2012;18(1):117-141. DOI: 10.1111/j.1468-036X.2010.00546.x
28. Wadhwa K., Neupane S., Syamala S. R. Do group-affiliated firms time their equity offerings? Pacific-Basin Finance Journal. 2019;54:73-92. DOI: 10.1016/j.pacfin.2019.02.004
29. Gompers P. A., Lerner J. The really long-run performance of initial public offerings: The pre-Nasdaq evidence. The Journal of Finance. 2003;58(4):1355-1392. DOI: 10.1111/1540-6261.00570
30. Ritter J. R., Welch I. A review of IPO activity, pricing, and allocations. The Journal of Finance. 2002;57(4):1795-1828. DOI: 10.1111/1540-6261.00478
31. Fernando C. S., Krishnamurthy S., Spindt P. A. Are share price levels informative? Evidence from the ownership, pricing, turnover and performance of IPO firms. Journal of Financial Markets. 2004;7(4):377-403. DOI: 10.1016/j.finmar.2004.01.001
32. Habib M. A., Ljungqvist A. P. Underpricing and entrepreneurial wealth losses in IPOs: Theory and evidence. The Review of Financial Studies. 2001;14(2):433-458. DOI: 10.1093/rfs/14.2.433
33. Gupta V., Singh S., Yadav S. S. Disaggregated IPO returns, economic uncertainty and the longrun performance of SME IPOs. International Journal of Emerging Markets. 2021. DOI: 10.1108/ IJOEM-09-2020-1098
34. Dhamija S., Arora R. K. The long-run performance of graded IPOs in the Indian capital market. Global Business Review. 2014;15(2):317-337. DOI: 10.1177/0972150914523597
ABOUT THE AUTHORS / ИНФОРМАЦИЯ ОБ АВТОРАХ
Parveen Siwach — Research Scholar, National Institute of Food Technology Entrepreneurship and Management, Kundli, India; PhD, Assist. Prof., Amity University, Punjab, India
Парвин Сивач — научный сотрудник, Национальный институт предпринимательства и менеджмента в области пищевых технологий, Кундли, Индия; PhD, доцент, Университет Амити, Пенджаб, Индия https://orcid.org/0000-0002-2770-2251 siwach.parveen23@gmail.com
Prasanth R. Kumar — PhD, National Institute of Food Technology Entrepreneurship and Management, Kundli, India; Assoc. Prof., University of Hyderabad, Hyderabad, India Прасант Р. Кумар — PhD, Национальный институт предпринимательства и менеджмента в области пищевых технологий, Кундли, Индия; доцент, Университет Хайдарабада, Хайдарабад, Индия https://orcid.org/0000-0001-5299-7701 Corresponding author / Автор для корреспонденции: prasanth@uohyd.ac.in
Vikas Gupta — PhD, Assist. Prof., O. P. Jindal Global University, Sonipat, India
Викас Гупта — PhD, доцент, Глобальный университет имени О. П. Джиндала, Сонипат,
Индия
https://orcid.org/0000-0002-5601-8432 vgupta@jgu.edu.in
Authors' declared contribution:
P. Siwach — Problem statement, literature review, data analysis, interpretation, conclusion, and references. P. R. Kumar — Idea formation, Proof-reading, results discussion, conclusion, manuscript improvements. V. Gupta — Proof-reading, results interpretation, literature suggestions, manuscript improvements.
Заявленный вклад авторов:
П. Сивач — постановка проблемы, обзор литературы, анализ данных, интерпретация, заключение и ссылки.
П. Р. Кумар — формирование идеи, корректура, обсуждение результатов, заключение, доработка рукописи.
В. Гупта — корректура, интерпретация результатов, предложения по литературе, доработка рукописи.
Conflicts of Interest Statement: The authors have no conflicts of interest to declare. Конфликт интересов: авторы заявляют об отсутствии конфликта интересов.
The article was submitted on 28.09.2022; revised on 21.10.2022 and accepted for publication on 26.10.2022. The authors read and approved the final version of the manuscript.
Статья поступила в редакцию 28.09.2022; после рецензирования 21.10.2022; принята к публикации 26.10.2022.
Авторы прочитали и одобрили окончательный вариант рукописи.
