Presidential elections and stock market outcomes: An event-study on the effect of Turkey's Presidential Elections on Borsa Istanbul

Abstract

This paper uses the event-study methodology to investigate the effect of the Presidential Turkish elections in 2023 on Borsa Istanbul returns. The data used in this study cover the period from 13 June 2022, through 7 June 2023. We employ a market model to study the effect of two election rounds on the stock market’s cumulative abnormal returns (CARs). Our results show that the impact of the first round of the elections on the stock market is mixed, as it has a positive effect on CARs for event windows [−2, 2] and a negative impact on CARs for event windows [−6, 6] and [−7, 7]. However, following round 2 of the elections, results show that re-elections have a significant positive impact on CARs, with average CARs ranging from 678 basis point to 1019 basis point, indicating that uncertainty in the market has vanished and perhaps investors are optimistic about the future of Erdogan reign and his expected economic policies. Results also show that sectors’ returns reacted differently to the first round of the elections and round 2 of the re-elections. For example, the event has a significant negative effect on most of the sectors’ returns during the first presidential elections. In contrast, it has a significant positive impact on most of the sectors’ returns during round 2 re-elections.

Keywords:

• Erdogan
• Turkey
• Borsa Istanbul
• stock-market returns
• presidential elections

JEL Classification:

• G10
• G14
• G15

1. Introduction

Presidential elections are a crucial turning point in the history of any country because they commonly lead to changes in the administration and policy that can significantly affect the economic environment of any country. The stock market is one of the potential aspects that might be affected.

This Turkish election in 2023 takes place roughly 100 years after Mustafa Kemal Atatürk’s secular republic was established. If Erdogan triumphs, he will have even more authority to influence the course of 85 million people. In the West, there is concern that he may use this as an opportunity to advance a more regionally hostile, religiously conservative model and cantered on himself. The election will have a significant impact on Middle Eastern and European security. According to Stamouli (2023), whoever wins the election will determine Turkey’s character in NATO, its ties to the US, the EU, and Russia, its immigration policies, Ankara’s involvement in the Ukraine conflict, and how it manages tensions in the Eastern Mediterranean.

In Turkey, there are two rounds in the presidential election. Usually, many candidates run for the presidency in the first round, the unpredictability of the results might make the market unstable and increase the uncertainty (Asteriou & Sarantidis, 2016; MengYun et al., 2018). If no candidate receives a simple majority after the first round, the election moves on to a runoff between the two front-runners. Campaigning is normally heating up at this time, and we believe market returns may also be affected because the range of possible candidates who may win has decreased to two candidates.

The candidates who ran for the presidency in the first round on Sunday, 14 May 2023, were Turkish President Recep Tayyip Erdogan, who also serves as the head of the Justice and Development Party, Kemal Klçdarolu, the Republican People’s Party head, and leading opponent, former Good Party head and the independent nominee Muharrem ince, and former Nationalist Movement Party member of parliament and independent nominee Sinan Ogan. As discussed by Tuysuz et al. (2023), no candidate obtained a majority of the vote and a runoff election between Erdogan and Klçdarolu was held on Sunday, 28 May 2023. With 52.14% of the vote, Erdogan won the runoff election and was re-elected as Turkey’s president, while the other candidate received 47.86% of the votes.

The Turkish economy has suffered recently from weak economic conditions, with an inflation rate that has increased to double digits, reaching 85% in October 2022 (Sevgili et al., 2023). Recep Tayyip Erdogan, the president of Turkey, has a very unorthodox perspective on interest rates. Instead of the other way around, he thinks that high interest rates cause inflation. Based on his Islamic beliefs, Erdogan has an opinion about interest rates. Islam forbids interest, often known as “Riba.” In Erdogan’s opinion, high rates on interest are a type of riba and are detrimental to the economy (Kutlay, 2021). His views contradict economists’ belief that high interest rates help keep inflation under control (Dmitracova, 2023). Therefore, this has led central bank officials not to raise interest rates in a way to lower inflation (Dmitracova, 2023). Instead, the Turkish government has implemented price restrictions steps to curb inflation (Dailysabah, 2023; Yermolenko, 2023).

Unfortunately, the Turkish government’s price restrictions steps have not been very practical, though, so far, and as a result, the Turkish lira’s value has depreciated against the major currencies (Sevgili & George, 2023). The Turkish government’s ability to control inflation in the upcoming years is still challenging and uncertain.

We believe that this paper comes with a unique case in a very different time for a very different country, given that Erdogan’s role and policies nowadays are being observed by many countries as the media is calling Erdogan the new Sultan Erdogan (Khalid, 2023). The consequences of this election will have a significant impact on Turkey’s relations with the European Union and the United States of America and the country’s future. In addition, the Turkish economy is struggling this time with a high inflation rate and the low value of the currency against the US dollar, which has led to great dissatisfaction among the Turkish people. Therefore, we have an exclusive case where two events occur in a very short period, that is, round 1 elections and round 2 re-elections in only two weeks gap. We examine the effect of election results on stock market outcomes. In particular, we examine the effects of round 1 elections when there was no winner and round 2 re-elections—the triumphs of Erdogan—on stock market returns. To the authors, this is the first paper covering this unique case.

Our results show that during round 1, where the result is not clear, the uncertainty is at its highest. The results are not significant for most of the event windows. In contrast, at round 2, where the result becomes apparent, the uncertainty vanishes as the market reacts positively to the event and we get significant positive results for all the event window. This paper sheds light and adds to the literature examining the relationship between presidential elections and stock market outcomes in an emerging market.

The rest of this paper is structured as follows: the literature review in section 2 and the data and methodology in section 3. The results and analysis are presented in section 4. Further test for the sectors in section 5. The paper is concluded in section 6.

2. Literature review

Over the past several years, there has been considerable discussion over how political elections and political interventions in central bank policies affect the performance of the financial markets and stock market outcomes (Bash & Al-Awadhi, 2023). Using an event study design, Afego et al. (2023) measured abnormal returns prior to and following the announcement of the election results in Nigeria. They found that the market responded positively in the 2015 elections when there was a change in the presidency to an opposition party candidate. However, in the re-elections in 2019, the response was weakly positive. Regarding this, Afego et al. (2023), like Harrington (1993), concluded that a first-term president’s chances for re-election can be affected directly by the policies he chooses and indirectly by how those policies affect economic performance.

In another study, Mnasri and Essaddam (2021) empirically investigated the volatility of the US stock returns during election windows. Their research showed that stock volatility is affected by political uncertainty through macroeconomic channels. Similarly, Musah et al. (2023) examined the effect of presidential elections on stock return volatility in five major sub-Saharan African stock markets. The results of their study reflected that the Nairobi Stock Exchange, the Stock Exchange of Mauritius, and the Nigerian Stock Exchange all exhibited more volatility when there was substantial pre-election uncertainty. Furthermore, their findings demonstrated that stock return volatility decreased 90 days after elections in South Africa and Nigeria but increased 90 days after elections in Ghana.

In a similar study, Shen et al. (2017) studied how political aspects such as government policies and political connections affect stock returns by focusing on the 2008 Taiwanese presidential election. They discovered that, companies that benefited from (were endangered by) the winning party’s planned Three-Links policy achieved positive (negative) stock returns during the election. To examine the impact of government policy, Shen et al. (2017) looked at how sensitive company returns are to bilateral trade flows between Taiwan and China. According to their findings, the effects of political connections are only marginally present, but they become more substantial as the winning party’s support ratio rises in polling data. Additionally, Shen et al. (2017) discovered that only the government-policy effect is consistent across different levels of corporate governance and crash risk. The authors concluded that based on the Fama French Three-Factor Model, investment techniques focused on both political aspects can produce positive abnormal returns. Santa-Clara and Valkanov (2003) in a study on whether Republicans or Democrats are better for the stock market found that average excess returns under Democrats are higher than under Republican presidency due to business cycle variables related to expected returns.

According to Anderson et al. (2023), the relationship between political events and stock markets has long been under concern as long as there is interest in market efficiency. On this account, the research by Kolaric and Schiereck (2016) shows that in an efficient market, stock prices are anticipated to change instantly and without overreacting, and the impact of an event is anticipated to fade over time if the event occurs repeatedly. Further, to understand the reaction of stock market outcomes toward political elections, Pantzalis et al. (2000) also investigated how stock market indexes behaved around election dates in 33 countries between 1974 and 1995. They discovered a positive abnormal return in the two weeks preceding election week. The authors concluded that the positive abnormal returns were influenced by a country’s level of political, economic, and press freedom, as well as the timing of the election and the effectiveness of the incumbent in winning re-election.

By employing an event-study technique, Antoniuk and Leirvik (2021) examined how unanticipated political events affect industries sensitive to climate change. They discover that returns have been greatly affected by events linked to climate change policy. The Paris Agreement, Climategate, and Fukushima all positively affected the clean energy sector because they raised public awareness of climate change and supported legislation aimed to reduce its effects. Events linked to climate change policy indicate heightened political and market risks, which require more compensation for the utilities, energy-intensive, and transport sectors. Positive abnormal returns for the fossil energy industry are correlated with events weakening climate change policies. They also discover that investors in the stock market are quick to adjust to new information about climate change. Because investors are likely to price in both climate risk and expectations for sectors’ growth, policymakers should be mindful of how such events affect the stock market.

The existing literature on the relationship between political elections and stock market outcomes shows that the former tends to cause a significant shift in stock market returns. However, the literature fails to distinguish the specific aspects of presidential elections that result to positive stock market returns and which cause a negative stock market outcome. Therefore, there is a need to conduct further research on the impact of a first round of presidential election on stock market returns and the change observed after a second of round re-election. Further research on this will, therefore significantly contribute to the existing literature by filling the gap.

3. Data and methodology

3.1. Data

From 13 June 2022, through 7 June 2023, daily data from Bloomberg were collected for our sample. When the election result is announced over the weekend, that is, Sunday, the first trading day of the following week, is taken into account as the event day. This means that Monday is used as our event day. Therefore, event 1 represents the round 1 election results announcement dated Monday, 15 May 2023, and event 2 represents the round 2 re-election results announcement dated Monday, 29 May 2023. The data contain the adjusted-closing prices of the Borsa Istanbul 100 Index (BIST100) and its constituents.

3.2. Methodology

We employ the event-study technique. Event studies use data from financial markets to determine how a particular event may affect a firm’s stock price. The advantage of this technique originates from the belief that the effect of an event is instantly reflected in stock prices in a rational market. As a result, Mackinlay (1997) asserts that it is possible to determine the event’s economic influence using security prices observed over a short period of time.

Daily return is calculated as follows:

(1) $D{R}_n=ln\left(\frac{P_{n,d}}{P_{n,d-1}}\right)$(1)

where $P_{n,d}$ is the price of stock; n at time d, and $P_{n,d-1}$is the price of stock; n at time $d-1$. $ln$ represents the natural logarithm, and $D{R}_n$ is the daily rate of return for stock n.

The market model is used to determine the daily abnormal returns (AR) and cumulative abnormal returns (CAR) for the observation period, including Day 0 of the event.

We follow Dodd and Warner (1983) and Brown and Warner (1985) and calculate abnormal returns as follows:

(2) $A{R}_{n,d}=D{R}_{n,d}-\left({\alpha }_n+{\beta }_{n}D{R}_{m,d}\right)$(2)

where $D{R}_{n,d}$ denotes the daily return for stock n at time d, as shown in Equation 4, $D{R}_{m,d}$ represents the daily return of the BIST100 index; the dependent variable $A{R}_{n,d}$is the daily abnormal returns for stock n at time d; $CA{R}_{n,d}$ is simply the summation of daily $AR$s for all the days in the event window; and ${\alpha }_n$ and ${\beta }_n$ denote the regression coefficient estimates obtained from the ordinary least squares method for the period [−250, −11] days.

4. Results and analysis

Figures 1 and 2 show abnormal returns AR and cumulative abnormal returns CAR surrounding event 1 and event 2. Figure 1 shows that CAR increased for two days following event 1 (15 May 2023) then declined. Figure 2 shows that CARs increased following event 2 (29 May 2023) and maintained its upward trend for about a week.

Figure 1. $\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ and $\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ for all stocks surrounding round 1 elections.

Figure 2. $\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ and $\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ for all stocks surrounding round 2 re-elections.

Tables 1 and 2 show descriptive statistics of AR and CAR for event 1 and event 2. Table 1 shows mean and median of AR and CAR, which is negative on event day. Table 2 shows mean and median of AR and CAR, which is positive on event day.

Table 1. Descriptive statistics of abnormal returns $\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ and cumulative abnormal returns $\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ before and after round 1 elections

		AR				CAR
t	Date	Mean	Median	Skewness	Kurtosis	Mean	Median	Skewness	Kurtosis
−10	5/5/23	−0.0017	−0.0017	0.8799	4.3345	−0.0017	−0.0017	0.8799	4.3345
−9	6/5/23	−0.0326	−0.0374	0.8178	2.4399	−0.0344	−0.0394	0.7930	2.2247
−8	7/5/23	−0.0256	−0.0250	−0.0296	−0.1659	−0.0600	−0.0592	−0.0342	−0.4316
−7	8/5/23	0.0070	0.0077	−0.0718	5.1576	−0.0530	−0.0525	0.0607	−0.2229
−6	9/5/23	−0.0126	−0.0125	−1.0225	3.7188	−0.0656	−0.0620	−0.0043	0.8396
−5	10/5/23	0.0143	0.0122	−0.3768	1.9134	−0.0513	−0.0566	−0.4975	1.3289
−4	11/5/23	−0.0116	−0.0155	0.8160	1.7494	−0.0630	−0.0669	−0.2410	0.7397
−3	12/5/23	−0.0121	−0.0119	0.5380	2.1012	−0.0751	−0.0787	0.0089	0.2243
−2	13/5/23	0.0312	0.0326	−0.3301	−0.3895	−0.0439	−0.0490	0.1136	−0.2932
−1	14/5/23	−0.0145	−0.0195	1.2331	2.5665	−0.0584	−0.0636	0.3454	−0.0567
0	15/5/23	−0.0140	−0.0148	0.5510	0.5229	−0.0724	−0.0711	0.0355	−0.0076
+1	16/5/23	0.0210	0.0269	−1.2906	2.5800	−0.0514	−0.0296	−0.4294	−0.4269
+2	17/5/23	0.0181	0.0153	0.6370	0.5052	−0.0332	−0.0373	−0.1551	−0.3837
+3	18/5/23	−0.0221	−0.0213	0.0181	1.4416	−0.0553	−0.0508	−0.1774	−0.2889
+4	19/5/23	−0.0017	−0.0017	0.8799	4.3345	−0.0571	−0.0531	−0.1778	−0.2978
+5	22/5/23	−0.0056	−0.0117	1.4231	2.4253	−0.0627	−0.0638	0.0785	−0.0864
+6	23/5/23	−0.0014	−0.0035	1.0445	3.0261	−0.0640	−0.0643	0.0467	−0.1396
+7	24/5/23	−0.0037	−0.0085	1.8434	4.9527	−0.0677	−0.0696	0.1935	0.363

Table 2. Descriptive statistics of abnormal returns $\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ and cumulative abnormal returns $\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ before and after round 2 re-elections

		AR				CAR
t	Date	Mean	Median	Skewness	Kurtosis	Mean	Median	Skewness	Kurtosis
−10	15/5/23	−0.0134	−0.0145	0.5576	0.5524	−0.0134	−0.0145	0.5576	0.5524
−9	16/5/23	0.0207	0.0266	−1.2796	2.5577	0.0073	0.0121	−0.4097	1.1599
−8	17/5/23	0.0179	0.0148	0.6454	0.5214	0.0253	0.0310	−0.2007	1.8185
−7	18/5/23	−0.0218	−0.0212	0.0073	1.4599	0.0035	0.0099	−0.6069	1.8810
−6	19/5/23	−0.0018	−0.0018	0.5565	1.1402	0.0017	0.0075	−0.5990	1.8705
−5	22/5/23	−0.0056	−0.0116	1.4314	2.4549	−0.0039	−0.0042	−0.0973	1.7373
−4	23/5/23	−0.0014	−0.0032	1.0420	3.0237	−0.0053	−0.0065	−0.2535	1.5437
−3	24/5/23	−0.0036	−0.0083	1.8409	4.9487	−0.0090	−0.0032	0.0097	1.9073
−2	25/5/23	0.0046	−0.0005	1.6488	4.4983	−0.0043	−0.0038	0.2366	1.5905
−1	26/5/23	0.0166	0.0154	0.8246	1.5193	0.0122	0.0150	0.2218	1.5375
0	29/5/23	0.0289	0.0232	0.4228	0.4113	0.0412	0.0493	0.0444	1.3945
+1	30/5/23	0.0223	0.0178	0.3820	0.1561	0.0635	0.0683	0.0663	1.2996
+2	31/5/23	0.0008	−0.0042	0.9499	1.2027	0.0643	0.0680	0.1948	1.1907
+3	1/6/23	0.0106	0.0089	0.4861	1.0778	0.0749	0.0769	0.1403	1.2033
+4	2/6/23	0.0068	0.0036	1.1379	2.3289	0.0817	0.0791	0.2147	1.7053
+5	5/6/23	0.0155	0.0119	0.8400	1.5977	0.0973	0.0897	0.1502	1.2331
+6	6/6/23	−0.0046	−0.0078	1.4609	4.7191	0.0926	0.0804	0.4552	2.6676
+7	7/6/23	0.0093	0.0063	0.0168	3.1212	0.1019	0.1035	0.1370	1.3788

Tables 3 and 4 show the mean and median equality tests for CAR using market model. Upon event 1, as shown in Table 3, most of the CARs were insignificant with the exception of event window [−2, 2], which shows the effect was significantly positive and event window [−7, 7], which shows the effect was significantly negative. This suggest that the first round of Turkish elections drew a mixed response from the stock market and implies that although there was no clear winner in the election, investors were initially apprehensive about how it would turn out. It is hypothesized that the uncertainty surrounding the election result has led investors to become uncertain about the future course of Turkish politics and the economy because there was no obvious winner. It is also worth to mention that as discussed by Cooban (2023), the Turkish lira has slipped to a new record low against the US dollar on the election date which could also add to this mixed result.

Table 3. Mean and median equality tests for$\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ – round 1 elections

Market Model
Event Window	Mean	t-test	Median	W/M
[−1,1]	−0.0075	−0.9676	−0.0104	−1.3770
[−2,2]	0.0419	4.7043***	0.0370	4.4475***
[−3,3]	0.0076	0.9650	−0.0002	−0.5381
[−4,4]	−0.0057	−0.7925	−0.0134	−1.2154
[−5,5]	0.0030	0.3235	−0.0070	−0.1289
[−6,6]	−0.0110	−1.3479	−0.0238	−1.8481**
[−7,7]	−0.0677	−7.6193***	−0.0696	−6.3386***

Note. The test suggests a standard test for equality of the Satterthwaite–Welch; *, **, and *** suggest statistical significance at 10%, 5%, and 1% levels, respectively. W/M means Wilcoxon–Mann–Whitney signed-rank test.

Table 4. Mean and median equality tests for $\mathit{C}\mathit{A}{\mathit{R}}_{\mathit{n},\mathit{d}}$ – round 2 re-elections

Market Model
Event Window	Mean	t-test	Median	W/M
[−1,1]	0.0678	13.6515***	0.0638	8.2399***
[−2,2]	0.0733	11.9659***	0.0700	8.1093***
[−3,3]	0.0802	11.7503***	0.0749	8.0955***
[−4,4]	0.0856	12.0277***	0.0846	7.9993***
[−5,5]	0.0956	12.2356***	0.0900	8.0336***
[−6,6]	0.0892	11.0037***	0.0822	7.8858***
[−7,7]	0.1019	7.8062***	0.1035	6.4830***

Note. The test suggests a standard test for equality of the Satterthwaite–Welch; *, **, and *** suggest statistical significance at 10%, 5%, and 1% levels, respectively. W/M means the Wilcoxon–Mann–Whitney signed-rank test.

On the other hand, Table 4 shows that for all the event windows in event 2, the event has a significant positive effect on CARs. Average CARs ranges from 678 basis point to 1019 basis point. Our results indicate that uncertainty has been minimized and investors may view Recep Tayyip Erdogan as a reliable leader and perhaps the investors becoming optimistic about his intended future economic policies. We hope that this time, Erdogan may attempt to repeat the great economic history of growth and development during his reign especially for the period from 2002 to 2013 which was known as Turkish economic miracle.¹

It is also worth to mention that following the re-elections, Erdogan has named Mehmet Simsek as the new finance minister in his cabinet and the market reacted positively on that as BIST100 index increased by 3.03% following his appointment (Sevgili, 2023). Simsek is well known as a market-friendly policy man as he will implement Orthodox economic policies (Shan, 2023). In addition, Erdogan has also appointed former Hafize Erkan as the new governor for the Central Bank of Turkey. Mrs Erkan was a former experienced executive in Goldman Sachs and the First Republic Bank, and her appointment sends a positive signal to the investors as she will take further steps in collaboration with the finance minister to fix the Turkish economy and overcome current challenges.

5. Further test: Sector analysis

During such political events, we expect sectors to react differently to these events. Therefore, we investigate the reactions of round 1 elections and round 2 of re-elections on sectors’ returns by performing a panel data regression for the returns of the constituents of BIST100 representing the following sectors²: industrials, utilities, financials, materials, real estate, consumer discretionary, information technology, consumer staples, health care, energy, and communication services. For the returns of the stocks in each sector, we use dummy variable that take the value of 1 in the event day (D0), and zero otherwise.

Our results show a significant difference in returns between the sectors. Table 5 shows that during round 1 elections, the event had a significant negative effect on most of the sectors’ returns which are industrials, utilities, financials, consumer discretionary, consumer staples, health care, and energy, whereas it had insignificant effect on materials, real estate, information technology. Only communication services sector’s returns are significantly and positively affected by the event. On the other hand, results reveal that round 2 re-elections has a significant positive effect on all the sectors’ returns, except for information technology, energy, and communication services sectors’ returns where the effect has insignificant effect. These results are in general, consistent with our prior results using the event-study methodology.

Table 5. Sectors analysis

	Round 1 elections		Round 2 re-elections
	Coff.	D0	Coff.	D0
Industrials	0.0039***	−0.0547***	0.0036***	0.0455***
	(0.0004)	(0.0081)	(0.0004)	(0.0081)
Utilities	0.0032***	−0.0407***	0.0030***	0.0459***
	(0.0007)	(0.0123)	(0.0007)	(0.0123)
Financials	0.0039***	−0.0751***	0.0036***	0.0229**
	(0.0006)	(0.0106)	(0.0006)	(0.0107)
Materials	0.0032***	−0.0114	0.0030***	0.0625***
	(0.0005)	(0.0087)	(0.0005)	(0.0087)
Real Estate	0.0006	0.0230	0.0006	0.0306**
	(0.0009)	(0.0154)	(0.0009)	(0.0154)
Consumer Discretionary	0.0029***	−0.0453***	0.0026***	0.0501***
	(0.0005)	(0.0084)	(0.0005)	(0.0084)
Information Technology	0.0081***	−0.0258	0.0080***	0.0189
	(0.0028)	(0.0507)	(0.0028)	(0.0508)
Consumer Staples	0.0036***	−0.0792***	0.0033***	0.0187**
	(0.0006)	(0.0103)	(0.0006)	(0.0105)
Health Care	0.0033***	−0.0131	0.0032***	0.0489**
	(0.0011)	(0.0199)	(0.0011)	(0.0198)
Energy	0.0027**	−0.0545**	0.0024*	0.0317
	(0.0013)	(0.0226)	(0.0013)	(0.0227)
Communication Services	−0.0746*	0.0023***	0.0019	0.0383
	(0.0232)	(0.0013)	(0.0013)	(0.0233)

Note. *, **, and *** suggest statistical significance at 10%, 5%, and 1% levels, respectively.

6. Conclusion

This paper investigates the effect of the Presidential Turkish elections of 2023 on Borsa Istanbul returns using the event-study methodology. The data used in this study cover the period from 13 June 2022, through 7 June 2023. We employ a market model to study the effect of two election rounds on stock market’s cumulative abnormal returns. Our results show that the impact of the first round of the election on the stock market is mixed as it has a positive effect on CARs for event windows [−2, 2] and a negative effect on CARs for event windows [−6, 6] and [−7, 7]. However, following round 2 of the elections, results show that the re-elections have a significant positive impact on CARs with average CARs ranging from 678 basis point to 1019 basis point, indicating that uncertainty in the market has been minimized and investors are optimistic about the future of Erdogan’s reign and his expected economic policies. Results also show that sectors’ returns reacted differently to the first round of the elections and round 2 of the re-elections. For example, the event has a significant negative effect on most of the sectors’ returns during the first presidential elections whereas it has a significant positive effect on most of the sectors’ returns during round 2 re-elections.

This paper fills the gap in literature by studying the impact of the elections on stock market returns. It provides insights to investors on how to read the behaviour of the market and diversify their portfolios during the elections in an emerging market such as Turkey. In addition, firms doing business in Turkey should be aware of the consequences of the presidential elections on their business philosophies. This is because sometimes, the winning president may implement government policies that may either match or mismatch the business philosophy of the firms. For future research, we suggest researchers to study the interactions between stock-market outcomes and the Turkish Lira during presidential elections.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. For more on Turkish economic miracle, see https://www.reuters.com/world/middle-east/when-erdogans-turkish-economic-miracle-began-failing-2021-07-15/

2. These sectors are defined according to Industry Classification of Global Industry Classification Standard (GICS).