Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company

1. Introduction

Investment decisions in shipping for newbuildings are predominantly stratified into replacement, expansion and new entrance opportunities.  As far as replacement investment is concerned, it includes the allotment of capital for substituting vessels that no longer fulfil the company’s requisites and are, thus, available for demolition. Vessels are mainly demolished due to technical obsolescence, age, poor overall condition, specific market conditions, implementing new international regulations or following company policies. Secondly, expansionary investment decisions comprise of capital disbursement for actualizing the company’s growth strategy as a response to the present or upcoming market conditions. Lastly, the case of a new entrance investment embodies a substantial amount of capital that newcomers bring into the industry by purchasing newbuilding vessels. Companies decide to expand their fleet capacity primarily due to freight market conditions in order to maintain or increase their industry market share. (Engelen, Meersman, & van De Voorde, 2006), (Adland & Strandenes, 2007) (Stopford, 2009) (Greenwood & Hanson, 2015). The decision for a newbuilding order is immensely affected by the relation between the second-hand and the newbuilding prices (Merikas et al., 2008), (Stopford, 2009)due to lags in construction (Kalouptsidi, 2014) as usually shipping investors require instantaneous vessel delivery to benefit from high-level freight rates.

On the other hand, the decision to scrap a vessel, which is rather important as a capital-intensive asset will be permanently disposed, as previously mentioned is based on particular vessel characteristics, specific market conditions or due to implementation of international regulations. In general, higher scrapping levels are noticed for older vessels in poor condition for which both employment potential and capital appreciation scope are narrow. Furthermore, technical obsolescence usually increases running costs for vessels leading them earlier to the scrapyard. In addition, new regulations lead a vessel to a compulsory demolition whereas poor market conditions leave the option up to the owner to either operate the vessel at a loss, lay-up or scrap it. Nevertheless, while deciding to operate at a loss and lay-up are reversible choices depicting a hope for future profitability, the decision to scrap is permanent and that is why ship-owners historically incline to avoid it.  Scrapping is related to the freight rate second-hand and scrap market levels.

Buxton (1991) underlined that the economic rationale in operating a vessel or selling her in the second-hand market is little when both markets have decayed significantly. Knapp et al. (2008) proved an inverse connection between vessel profitability and the demolition probability, found a positive link between scrap prices and demolition probability, and detected no particular relation between flag, ownership or safety and scrapping. In relation to that, Alizadeh et al. in 2016 after investigating the capacity retirement in the drybulk sector confirmed both the aforementioned inverse relationship between earnings and demolition and the fact that increased demolition prices motivate scrapping activity. Another finding of the study was also that the probability for scrapping rises with freight volatility, interest rates and age (Alizadeh, et al., 2016). Finally, as Stopford states (2009) market expectations play a vital role under freight income uncertainty to shipping investment/divestment decisions and the real options theory should be adopted by investors in their decision making as it enhances the company flexibility (Dixit and Pindyck, 1994; Dikos, 2008; Gkochari, 2015; Kyriakou et al., 2017).

Investing or divesting decisions regarding vessels may often be the result of ship-owners altering their market perspective due to the actions of other peers, i.e. there is a degree of herd behaviour.

As it is stated in Hwang and Salmon’s article (2004) “herding arises when investors decide to imitate the observed decisions of others or movements in the market rather than follow their own beliefs and information.” Due to this trading behaviour, investors might be guided to move in the same direction and, consequently, herding can result in asset prices differ from their fundamental values (Bikhchandani et al., 1992; Nofsinger and Sias, 1999). Hence, herd behaviour examination might lead to an understanding of its effect on asset values (Chang et al., 2000) as for instance, shipping investors could be benefited from profitable opportunities that might arise because of this group activity. Moreover, Scharfstein and Stein (1990) argued that despite that the classical economic theory proposes that investment decisions are made by utilizing all the available information efficiently, group psychology often drives investments and therefore undermining the relation between information and market outcomes.

The objective of this paper is to examining some of the forces that may contribute in herd behaviour in the shipping markets.  The existent literature on herding is primarily focused on herding between institutional/retail investors (Sias, 2004; Kumar and Lee, 2006) or towards the market consensus consensus (e.g., Christie and Huang, 1995; Chang et al., 2000). Falling into the latter strand, this paper examines herding behaviour in both newbuilding and scrap markets of drybulk and tanker vessels.

To achieve the latter, an enhanced herding equation was used to examine whether investors decide based on shared common elements or they just copy the decisions of few distinguished investors because of an informational drawback. Herding in general was examined and decomposed into intentional and unintentional.  Thereupon, separately for each sector, asymmetric herd behaviour effects were tested regarding the contraction and expansion phases of the market and lastly whether there are any spill-over herding effects from one market to another.

Overall, the results show that shipping investors herd unintentionally in their decision to order or scrap in both the newbuilding and scrap markets and intentionally in decisions for scrapping. Moreover, there is evidence of asymmetric herding since herding is more likely to be present in the newbuilding market during profitable market conditions and in the scrap market during poor conditions; a finding not supported by Wald Test. Finally, spillover effects of unintentional herding are evident from the newbuilding market to scrapping.

This paper is structured as follows. Section 2 discusses total herding, how it is measured and detected , Section 3 includes the decomposing of herding into intentional and unintentional and Section 4 provides the result for the asymmetric effects of herding during the ups and downs of the market. Herding spill-over effects evidences are presented in Section 5 and finally Section 6 concludes this paper.

2. Herding Measurement and Detection

The fact that investors might overlook former heterogeneous beliefs in order to adopt correlated patterns around the market conduct is the insight behind the following herding measure (Christie & Huang, 1995) (Chang, et al., 2000). Thus, herding was investigated by using broadly two methods. In their article, Christie and Huang (1995) applied the cross-sectional standard deviation (CSSD) of stock returns to detect herding behaviour and based their empirical evaluation on rational asset pricing models and herding during bear market periods. They suggested that rational asset pricing models foresee that as the absolute values of market returns increase so does the cross-sectional returns dispersion because of the fact that investors’ trading behaviour is based on private and diverse information. Nevertheless, in times of market stress, lower return dispersions are realised due to the tendency of investors to imitate collective actions in the market disregarding their private information. Hence, herding is more extensive in extreme market movements that result in aberrant returns on the market portfolio. To tailor the assumptions to the appropriate scenarios, they detached the upper and lower tails of the returns distribution and investigated whether these deviate from the average level of dispersion.

Furthermore, herding can be tested by examining the cross-sectional absolute deviaton (CSAD) of stock returns that Chang et al. proposed (2000). In their article, herding is regarded as a function of the dispersion variable which is either non-linearly decreasing or peaks at a certain value of the expected absolute market return and descends thenceforth. This paper examines the existence of herding in the bulker and tanker sector by utilizing an adjusted CSAD for both contracting and scrapping markets. Hence, the CSAD used for this purpose incorporates the number of the vessels that have been demolished and the new vessels that have been contracted instead of the return on the market and the asset. The equation used for the analysis is the following:

                              =1 ∑ =1 | , − |,  { , }                ( 1 )

Where CSADtθ is the cross-sectional absolute deviation of contracting ( θ=C) and scrapping ( θ = S) of vessels, , is the number of vessels in the ith sector of the market that are contracted or scrapped at time and = ∑ =14 , 4 is the cross-sectional average number of vessels contracted or scrapped.

As it concerns the dry-bulk market, in the equation 1 i represents the capesize, panamax, handymax or handysize sectors and C was estimated by using  data provided by Clarksons Shipping Intelligence Network for the period January 1996 to May 2019.  

The development of C  for bulker contracting and scrapping in the examined period is presented in Figure 1. Generally, the CSAD measure for contracting can be considered comparatively constant with three outstanding oddities where the values deviate fairly from the market consensus. Firstly and more evidently, there is the peak of all-time high for the dry-bulk market in 2007-2008, next there is 2010 where investors assumed that the market would boom again and lastly in 2013-2014 again there was a positive sentiment among the market participants for the future. On the contrary, CSAD s is characterized by great volatility and big fluctuations from the average market behaviour can be detected.  These include:  the Russian and Asian crises (1996-1999), the internet bubble (2000-2002), the financial crisis of 2008 and the events that followed it and lastly the monumental freight dives during the period 2011-2013.

Figure 1: Cross-sectional absolute deviation ( ) for contracted and scrapped vessels in the bulker market (1996-2019)

As it concerns the dry-bulk market, in the equation 1 i represents the U/VLCC, suezmax or aframax sectors and   was estimated by using  data provided by Clarksons Shipping Intelligence Network for the period January 1996 to May 2019.  The development of C  for tanker contracting and scrapping in the examined period is presented in Figure 2. In general, C  is presented relatively stable with two outstanding deviations from the average. Firstly, the peak of the tanker market during 2006-2008 where it reached the highest level historically and secondly during 2013-2014 where the positive sentiment of the investors dominated the market.  On the other hand, CSADS shows noticeable volatility and deviates from the market consensus repeatedly. These include  the Asian crisis  (1998-1999), the internet  bubble (2000-2002) , the transition from single-hulled tanker vessels because of the Oil Pollution Act of 1990 when Europe banned these vessels(2003)and lastly due to new IMO regulations and continuously low freights 2018 which was a year with increased scrapping activity.

Figure 2: Cross-sectional absolute deviation ( ) for contracted and scrapped vessels in the tanker market (1996-2019)

Comparing the two markets, it can be concluded that investors in both markets act in a similar way. As it regards the contracting, while in both sectors the deviations from the market consensus exist in the same periods the tanker market seems more volatile. Investors decide to allocate capital to newbuilding vessels in times where the overall sentiment of the market is positive. On the other hand, as scrapping is concerned the picture is different. Besides the fact that dry-bulk and tanker owners decide to scrap their vessels in different periods triggered by different events, tanker scrapping deviations are also more frequent and erratic.

In order to examine if herding activity exists, the non-linear OLS specification that Chang et al. (2000) proposed is used and the link between the cross-sectional absolute deviation (CSAD) and the overall market average of vessel contracting or scrapping, respectively, was estimated[1].

= 0 + 1 | | + 2 ( )2 + , ( 2 )

Chang et al. (2000) indicated that, if investors are inclined to adopt the behaviour of the market during times of large price volatility, then the rational asset pricing models that indicate the linear and increasing relation between dispersion and return no longer hold and this relation might become non-linearly increasing or decreasing. The positive coefficient γ1 illustrates the linear part of the above relation, while the non-linear part by coefficient γ2. If γ2 is negative (γ2 < 0) , then the cross-sectional deviation of contracting or scrapping grows less than linearly or even declines, when the overall absolute market average is large. This is explained as evidence of existent herding behaviour and, therefore, the coefficient of the non-linear term should be negative and statistically significant. Thus, this paper assumes that if herding behaviour exists for either contracting or scrapping, then coefficient γ2 has to be negative and statistically significant.

3. Decomposing Herding into Intentional and Unintentional

Herding behaviour can be divided into intentional and unintentional herding. Regarding intentional herding, investors imitate other investors’ decisions with intent. This kind of conduct is ordinarily seen in less sophisticated investors who seek to mimic respectable or well-established investors, since acquiring the complete information of these market players would be costly. In general, intentional herd behaviour is defined by professional, as a result of ability or reputation, or informational asymmetry, when investors believe they have an informational disadvantage. (Devenow and Welch, 1996) This is frequently seen in hedge funds’ or financial intermediaries’ managers who are constantly evaluated and therefore less qualified managers may imitate the actions of the more sophisticated or more reputable peers. (Scharfstein and Stein, 1990; Villatoro, 2009).  In contrast, unintentional herd behaviour is observed when because of a sharing element in their environment investors come to similar investment decisions Hirshleifer et al., 1994; Bikhchandani and Sharma, 2000). Share elements may contain characteristic trading (Bennett et al., 2003) and relative homogeneity (Teh and de Bondt, 1997). Characteristic trading refers to investment decisions taken on the basis of certain characteristics of the assets, which in the course of time result in style investing (e.g income, momentum, growth, industry trading) whereas relative homogeneity refers to the way investors process financial information or signals (e.g. financial ratios) received from the market in a similar way because of the fact they share similar analytical skills or academic backgrounds (Wermers, 1999).

To decompose the cross-sectional absolute deviaton CSADϑ t into intentional and unintentional components, three parameters that sufficiently capture crucial shipping information, are similar to all market players and might influence the choice to order a new or demolish an old vessel were used. The fact that there are many other parameters affecting this decision is recognised nevertheless. These three parameters cover major areas of the shipping market and are divided into: valuation (price-earnings ratio) and market conditions (secondhand-newbuilding price ratio and Clarksea Index). All the required data that were used for the estimation of the variables are provided by by Clarksons Shipping Intelligence Network for the period January 1996–May 2019.

The first estimated parameter is the price-to-earnings ratio (PE) for vessels: PEi,t = PSH it − Eit, where PSHit is the log-price of the 5-year old secondhand vessel and Eit the log-earnings (Clarksea Earnings for bulkers and tankers respectively)[2] in sector i and month t. This ratio is applied to predict subsequent asset returns (Campbell & Shiller, 1998), (Rangvid, 2006), (Alizadeh & Nomikos, 2007) and points out the relative level of overvaluation or undervaluation in asset prices. This estimate is forwarded-looking and points out the expected earnings from an one year operation of the vessel from the point of valuation .i.e. high(low) P/E ratios are interpreted as high(low) present vessel prices compared to the one-year earnings. Papapostolou et al. (2014) in their article underline that high P/E ratios are linked with low investment sentiment levels which might lead to higher levels of demolition activity and low investment in contracting newbuilding vessels.

The second parameter, which is part of the market conditions category, is the secondhand-to-newbuilding price (SNB) ratio: SNBi,t = PSHit − PNBi,t , where PNB i,t is the log newbuilding vessel price. It need to be underlined that because newbuilding vessels have longer economic lives than similar secondhand vessels of certain age (e.g., five-year-old vessels), they are, in general, more expensive. However, it is important to realize that during bullish and profitable freight rate markets, investors in order to take advantage of these favourable conditions immediately prefer to purchase secondhand vessels to bypass the construction lags of newbuilding vessels. This leads to a delivery premium which drives the SNB ratio to higher levels. As Papapostolou et al. (2014)report , the SNB ratio is connected with investors sentiment for the shipping market and ,therefore, a higher ratio is expected to lead to less scrapping and more newbuilding orders.

Finally, the second parameter of the market conditions category is the Clarksea Index: C.I= Cit-Ct-12 where Cit is the log Clarksea Index level in month t.  Similarly to SNB ratio, it is expected that higher freight market rates can lead investors to order new vessels while scrapping remains at the minimum possible level.

To provide evidence that the aforementioned parameters contain useful information that may influence the decision to contract or to scrap vessels, the following regression is estimated:  Iϑt = β0 +β1Xt +υϑt ; where Iϑt =P4i=1Iϑi,t is the total number of vessels contracted or scrapped and Xt includes the aggregate metrics P/Et , SNBt and the CSIt.To calculate the aggregate metrics P/Et and SNBt for the drybulk and tanker market, weights ωi,t were assigned to each metric of sector i based on the market share (in terms of deadweight tonnage) of the sector in the total drybulk and tanker fleet respectively: PEt = P4i=1ωi,tP Ei,t and SNBt = P4i=1ωi,tSNBi,t. The results are presented in Appendix A and suggest that the proposed parameters explain a significant proportion of the variance of the number of vessels contracted or scrapped.

Finally, in order to decompose CSAD into intentional and unintentional herding the following regression is estimated:               = 0 + 1 + (3)

Intentional herding is defined as , = and unintentional herding is given by the difference between total herding and intentional herding

                                                   CSADϑ,Ut = CSADϑt − CSADϑ,I (4)

Thus, υϑt term can be defined as a measure of clustering due to the fact that investors respond to uncorrelated information, whereas CSADϑ,U t as a measure of clustering due to the fact that correlated information is analyzed in a similar manner.

Table 1:Herding behaviour in the drybulk market; intentional and unintentional herding

After decomposing the total herding behaviour into the unintentional and intentional derivatives, the regression model given by Equation (2) is calculated. The results for the drybulk market are illustrated by the OLS estimates (symbolised by γ0, γ1and γ2) in Table 1 and are classified in three different classes: total herding (CSADϑt), unintentional herding (CSADϑ,U,t) and intentional herding (CSADϑ,It).[3]In regard to total herding, the cross-sectional contracting and scrapping dispersions rise as the cross-sectional market average number of contracting and scrapping rises as well, a conclusion that is consistent with rational asset pricing models. However, herding activity is observed only in the scrap market as reflected by the negative and statistically significant value of γ2.  The results slightly differ qualitatively in the cases of unintentional and intentional herding. While, regarding unintentional herding, there is evidence of reduced cross-sectional dispersion around the market average both of contracting and scrapping, in the case of intentional herding this is observed only in the scrapping market. Consequently, it can be suggested that in the period January 1996-May 2019, investors, in the drybulk market, herded both intentionally and unintentionally regarding their decision to demolish older vessels but in the case of contracting new vessels they herded only unintentionally. Hence, scrapping herding behaviour is observed in every way (total, intentionally and unintentionally) and contracting herding behaviour is only observed unintentionally. The unconditional herding behaviour in scrapping and contracting markets can be associated with relative homogeneity. Particularly, these investment decisions were reached due to equivalent academic backgrounds and/or similar skills, which led investors to process the correlated information received in a similar way.

Table 2: Herding behaviour in the tanker market; intentional and unintentional herding

Accordingly, the results of the OLS estimates calculated by equation (2) for the tanker market are illustrated in Table 2. The results are quite similar for the two market sectors. In particular, in regard to total tanker herding the cross-sectional contracting and scrapping dispersions rise as the cross-sectional market average number of contracting and scrapping rises as well, a conclusion that, as previously mentioned, is consistent with rational asset pricing models. However, herding activity is, again, observed only in the scrap market as reflected by the negative and statistically significant value of γ2.  While, regarding unintentional tanker herding, there is again evidence of reduced cross-sectional dispersion around the market average both of contracting and scrapping, in the case of intentional herding this is observed only in the scrapping market. Therefore, it can be argued that in the period January 1996-May 2019, investors, in the tanker market, herded both intentionally and unintentionally regarding their decision to demolish older vessels but in the case of contracting new vessels they herded only unintentionally. Hence, scrapping herding behaviour is observed in every way (total, intentionally and unintentionally) and contracting herding behaviour is only observed unintentionally.

On the whole, shipping investors in the drybulk and the tanker sectors seem to herd in accordance. In both sectors relative homogeneity leads them, unintentionally, to take similar decisions in regards of contracting and scrapping. Of course, some less-sophisticated market players tend to intentionally mimic the decisions of well-established investors in the case of scrapping activity.

4. Asymmetric Herding during the Ups and Downs of the Market

In this section, the existence of an asymmetric relationship between CSADϑt and the cross-sectional market average of contracting and scrapping is examined. Researchers such as Christie and Huang (1995), Chang et al. (2000) and Demirer et al. (2010) argue that herding behaviour is expected to be more evident during downturn market periods. The assumption that shipping investors behave differently during periods when the freight rates are up compared to periods when  the freight rates are down, is tested by the approach  of Chiang and Zheng (2010) who use a single model with a dummy variable approach illustrated by :

= 0 + 1 (1 −  )| | + 2 | | + 3 (1 − )( )2 + 4 ( )2 + , ( 5 ) 

where d = 1 if CI,r < 0 (down market), and 0 otherwise (up market) for dry bulk and tanker markets.

Table 3: Herding behaviour under up and down dry bulk market conditions

	γ0	γ1	γ2	γ3	γ4	R2	Wald Test
Panel A: Total CSADtC CSADtS	1.5414α	0.2412α	0.2389α	0.0036γ
					0.0011	0.6716	[1.7127]
	(0.3405)	(0.0681)	(0.0423)	(0.0020)	(0.0008)
	0.2256	0.8196α  Share this: Facebook Twitter Reddit LinkedIn WhatsApp   Cite This Work To export a reference to this article please select a referencing stye below: APA MLA MLA-7 Harvard Vancouver Wikipedia OSCOLA UKEssays. (November 2018). Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company. Retrieved from https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1 Copy to Clipboard Reference Copied to Clipboard. "Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company." ukessays.com. 11 2018. UKEssays. 04 2024 <https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1>. Copy to Clipboard Reference Copied to Clipboard. "Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company." UKEssays. ukessays.com, November 2018. Web. 26 April 2024. <https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1>. Copy to Clipboard Reference Copied to Clipboard. UKEssays. November 2018. Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company. [online]. Available from: https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1 [Accessed 26 April 2024]. Copy to Clipboard Reference Copied to Clipboard. UKEssays. Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company [Internet]. November 2018. [Accessed 26 April 2024]; Available from: https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1. Copy to Clipboard Reference Copied to Clipboard. <ref>{{cite web|last=Answers |first=All |url=https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1 |title=Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company |publisher=UKEssays.com |date=November 2018 |accessdate=26 April 2024 |location=Nottingham, UK}}</ref> Copy to Clipboard Reference Copied to Clipboard. All Answers ltd, 'Herd Behaviour in the Drybulk and Tanker Marker: Decision to Invest in New and Retire in Fleet Company' (UKEssays.com, April 2024) <https://www.ukessays.com/assignments/herd-behaviour-in-the-drybulk-and-tanker-marker-decision-to-invest-new-fleet-company-2021.php?vref=1> accessed 26 April 2024 Copy to Clipboard Reference Copied to Clipboard. Related Services View all Assignment Writing Service From £124 Dissertation Writing Service From £124 Essay Writing Service From £124 DMCA / Removal Request If you are the original writer of this assignment and no longer wish to have your work published on UKEssays.com then please: Request the removal of this assignment Related Services Our academic writing and marking services can help you! Assignment Writing Service Prices from £124 Approximate costs for: Undergraduate 2:2 1000 words 7 day delivery Order an Assignment Dissertation Writing Service Essay Writing Service Marking Service Samples of our Service Full Service Portfolio Humanity University Dedicated to your worth and value as a human being! Learn More Related Lectures Study for free with our range of university lectures! All Available Lectures Freelance Writing Jobs Looking for a flexible role? Do you have a 2:1 degree or higher? Apply Today! Study Resources Free resources to assist you with your university studies! Dissertation Resources at UKDiss.com How to Write an Essay Essay Buyers Guide Referencing Tools Essay Writing Guides Masters Writing Guides Search Support Articles *You can also browse our support articles here > Close Change Region / Country Here you can choose which regional hub you wish to view, providing you with the most relevant information we have for your specific region. If your specific country is not listed, please select the UK version of the site, as this is best suited to international visitors. United Kingdom United States United Arab Emirates Saudi Arabia Bahrain Kuwait Oman Qatar Singapore Hong Kong Close