4.1. The Market Timing and the Selection Skills of Islamic and Conventional Funds
Table 5 reports the result for the five portfolios benchmarked each time against two different market indices. Following
Omri et al. (
2018) the performance of each funds is regressed against the relevant benchmark. Focusing on
Table 5, results indicate that 70% of funds operated by HSBC have significant statistics market selection.
The regression results show that the local Islamic portfolio (LS) have a statistically positive significant α. These results report the selectivity skills for local Sharia portfolio (LS) are lower when the portfolio is benchmarked against the S&P Saudi Arabia total return index (LCBM). Therefore, when benchmarked this portfolio with the S&P Quality Saudi Arabia Sharia Index (LSBM), we showed that the local Islamic portfolio have a significant and higher selectivity skills. This conclusion holds for the two different market benchmarks. On the other hand, local conventional portfolios (LC) generally do not reflect any evidence for a managers selectivity skills even when using two different market indices.
This finding indicates that market selection skills of local Islamic portfolio (LS) are stronger than the local conventional portfolio (LC). The managers of Saudi Islamic portfolios (LS) are able to take advantage of earnings anomalies, and thus, they make the right choice among stocks and sectors of activity. Our results specify that local Islamic funds are highly competitive and generate good returns with reasonable risk. The result may indicates the presence of home bias towards Islamic assets, due to the fact that manager or investor prefer invest in Saudi Islamic companies how prohibit alcohol, pork products, Gambling and any other forms whose activities the Sharia Board feels are prejudicial to the interests of Islam or Muslims.
We observed that the HSBC managers invested locally performed better in stock selectivity skill. The Islamic funds can able to select the ideal stocks for example, the stocks of firm who are not financially distressed, are growth oriented and are exhibiting a positive momentum.
Using two different benchmarks, S&P GLOBAL 1200 (INSBM) and S&P Global BMI Sharia (INCBM), we found that the international conventional portfolio (IC) have a good capacities to forecast security prices. This same table reports that if the Arab Islamic portfolios present negative and statistically significant α, then these managers are not doing well in funds selection.
Nevertheless, when the S&P Global benchmark Sharia or S&P GLOBAL 1200 (INSBM) are used as international market index, the results indicate that international Islamic portfolios (IS) have a higher statistically positive significant α than the international conventional portfolio (IC). The results also provide evidence of better performance in stock selectivity skill of Islamic portfolios (IS) comparing to the international conventional portfolios (IC).
An overall sample reveals a significant stock selectivity of the HSBC funds’ performance over the period of 2011 through 2018. Finally, 50% of the managers of these funds are able to take advantage of compensation anomalies and so they know how to make the right choice in the selection of securities and business sectors. These portfolios are able to generate higher excess returns than all the portfolios in the study.
The results of significant stock selectivity in Saudi Islamic portfolio are inconsistent with the finding of
Hayat and Kraeussl (
2011). The contrasting finding can be due to the shorter duration of the study period and the limited sample of Saudi mutual funds. However, the evidence of positive market selectivity of the fund managers supports the previous study by
Omri et al. (
2018), as they denoted that the Islamic funds are able to obtain superior return by investing in large, well-diversified corporations while applying the purification process through profits distribution.
In keeping with the approach of
Ferson and Schadt (
1996), positive and statistically significant β2 value indicate evidence of timing ability of the associated mutual funds. Negative significant β2 values may be interpreted as negative market timing.
Table 5 reports a significant market timing ability coefficient for only 40% HSBC portfolios (as well as international funds; IC and IS).
When the international-focused conventional portfolio (IC) is benchmarked against two different Index, the results report a positive and statistically significant coefficient of timing skills at 1 percent. We conclude that on average the managers of HSBC conventional funds investing internationally have little market timing ability and selection skills. In addition, the managers of this portfolio have little capacity to beat the market index by predicting its movements and buying and selling accordingly.
However, the results show that there are differences in market timing between the Islamic and the conventional internationally-focused portfolios. We showed that Islamic internationally-focused portfolios (IS) present negative and statistically significant coefficient regardless of the choice of the benchmark portfolio. This negative value may be interpreted as the incapacity of (IS) fund manager to beat the market. Moreover, the (IS) fund managers tend to augment (reduce) their exposures to the market in question at a time when the market is declining (rising). This evidence about the poor market timing is consistent with research conducted by
Fauziah et al. (
2002), who failed to capture the contribution of a manager’s timing activities to fund returns.
Furthermore, the results indicate evidence of differences in market timing skills between the Islamic and the conventional internationally-focused fund portfolios. The paper also provides evidence that the Conventional fund managers performing better in market timing ability (significant at 10%) in relation to the market performance. These results are consistent with the findings of research conducted by
Hoepner et al. (
2011) and
Hayat and Kraeussl (
2011). According to these authors, this insignificance of market timing can be recognized like the fact that the flow of Islamic funds is more stable than conventional funds, because on average the manager of Islamic funds select stock with long term perspectives and lower market risk.
Globally, the performance of the (IS) funds in stock selectivity skill is higher, and in market timing ability is slightly lower than its (IC) funds peers, with both statistically significant at 10%. Finally, it is evident that the investors are better off investing in international-focused conventional funds (IC) than international-focused Sharia funds (IS). This finding is due to the Sharia selection process and the way in which low-leverage firms are part of Islamic funds. This finding are inconsistent with the results of
Treynor and Mazuy (
1966) and
Girard and Hassan (
2008), who suggested that there are no differences in market timing skills between the Islamic and the conventional internationally-focused fund portfolios.
The inability to timing skills of the 60% other HSBC portfolios (LS; LC and AS) can be endorsed to the fact that the flow of local and Arab funds is more constant and persistent than HSBC international portfolios. Because investors in domestic and Arab funds are less sensitive to volatility of market returns, the manager chooses to invest in the long-term stocks.
Thus, regarding the first hypothesis, the results reveal that both Islamic and conventional portfolios have similar market timing and selectivity skills on the local and global and Arabic basis. Thus, the first hypothesis can be rejected, implying there is a statistically significant difference in terms of performance and strategy management between Islamic and conventional portfolios.
After that, we found strong evidence that the holy month of Ramadan has no affect on the market ability of HSBC portfolios, except on the international-focused conventional funds. The coefficient β4, which reflects the response of the portfolio manager’s beta to Ramadan effect is not statistically significant for 90% HSBC portfolios. According to these finding, no evidence was found to support the effect of Ramadan on the performance of the Islamic and conventional HSBC Saudi Fund.
For this reason, we can reject the second hypothesis, which was concerned with that the selectivity and market timing of HSBC Islamic portfolios are more sensitive to Ramadan effect than HSBC conventional portfolios. These results are inconsistent with the work of
Matthew et al. (
2014), who suggested evidence for a significant Ramadan effect within Muslim majority countries and regions.
To take an account for Heteroscedasticity in the error term, we used the GARCH model. According to Panel 2, we find that the parameters α’ and β’ of ARCH and GARCH are statistically significant and all most case, that indicate volatility clustering effect. Furthermore the sum of GARCH and ARCH parameter are near of one indicating volatility persistence.
4.2. Persistence of Performance
In order, to clarify the phenomenon of persistence of performance of HSBC portfolio, we have applied a measurement model, which inspired by
Grinblatt and Titman (
1992). This regression is basically used to find out if past performance is a good predictor of future performance. This regression is presented as follows:
where
represent the performance of each portfolio
p in the periods
t + 1 and
t, respectively. The hypotheses tested were:
H’0: 0
H’1: ≠ 0
Thus, if we accept the null hypothesis, there is no relationship between performance (t + 1) and performance (t).
Alternatively, if we reject hypothesis H’0, there is a relationship between the performance (t + 1) and the performance (t).
Table 6 shows the performance persistence of 15 HSBC portfolios funds over a period from 2011 to 2018, measured through the simple regression presented previously in Equation (4). By looking at the coefficient
we can conclude the performance of next month, as well as the persistence of performance.
According to the table below, we note that all portfolios managed by HSBC Saudi mutual funds present a coefficient different than zero; for this reason, we are able to reject hypothesis H’0 that implies that, there is a relationship between the performance (t + 1) and the performance (t). In spite of that this relationship does not give an answer about the existence of persistence.
From
Table 6 we can show that 50% of portfolios have a significant
coefficient at the 10 per cent level, as well as: international-focused conventional portfolio (IC), Arab sharia (AS) and Local Sharia (LS). Despite that, the international-focused conventional portfolio (IC) is the unique portfolio that has positive significant coefficients this means that, IC portfolio present a persistent performance during eight years. The managers of these portfolios have a certain degree of predictability of future performance, so they are able to maintain a certain stability in their performance. Moreover, the existence of performance persistence of (IC) confirms that the managers of this portfolio are encouraged to make their investment choices based on the historical performance. The performance persistence finding is considered as a test of manager’s market ability to underperform the market index. Finally, our empirical results reported in
Table 6 provide empirical evidence to reject hypothesis H3, because only a special case of conventional portfolios (as well as the IC portfolio) are able to present a stable performance and all Islamic portfolios are not able to perform significantly.