City, University of London Institutional Repository
Citation: Lim, K. G., Nomikos, N. ORCID: 0000-0003-1621-2991 and Yap, N. (2019).
Understanding the fundamentals of freight markets volatility. Transportation Research Part
E: Logistics and Transportation Review, 130, pp. 1-15. doi: 10.1016/j.tre.2019.08.003
This is the accepted version of the paper.
This version of the publication may differ from the final published
version.
Permanent repository link: https://openaccess.city.ac.uk/id/eprint/22733/
Link to published version: http://dx.doi.org/10.1016/j.tre.2019.08.003
Copyright: City Research Online aims to make research outputs of City,
University of London available to a wider audience. Copyright and Moral
Rights remain with the author(s) and/or copyright holders. URLs from
City Research Online may be freely distributed and linked to.
Reuse: Copies of full items can be used for personal research or study,
educational, or not-for-profit purposes without prior permission or
charge. Provided that the authors, title and full bibliographic details are
credited, a hyperlink and/or URL is given for the original metadata page
and the content is not changed in any way.
City Research Online: http://openaccess.city.ac.uk/ publications@city.ac.uk
City Research Online
Understanding the Fundamentals of Freight
Markets Volatility
Kian Guan Lim
a
, Nikos K Nomikos
b
and Nelson Yap
c
a
Lee Kong Chian School of Business, Singapore Management University, 50 Stamford Road, Singapore
178899. Email: kglim@smu.edu.sg.
b
Corresponding author: Cass Business School, Faculty of Finance, City, University of London, 106 Bunhill
Row, London EC1Y 8TZ, UK. Contact number: +44 (0)20 7040 0104. Email: N.Nomikos@city.ac.uk.
c
Lee Kong Chian School of Business, Singapore Management University, 50 Stamford Road, Singapore
178899. Email: nelson.yap.2013@pbs.smu.edu.sg
Abstract
We analyse empirically the drivers of freight market volatility. We use several macroeconomic
and shipping-related factors that are known to affect the supply and demand for shipping and
examine their impact on the term structure of freight options implied volatilities (IV). We find that
the level of IVs is affected by the level of the spot rate, the slope of the forward curve, as well as
by both demand and supply factors, especially the former. We demonstrate that the relation
between the volatility of futures prices and the slope of the forward curve is non-monotonic and
convex, that is, it has a V-shape. In general, anticipation of economic growth and of a stronger
freight market reduce IV whereas higher uncertainty and anticipation of excess shipping capacity
may increase IV. Panel regressions as well as a series of robustness tests produce strong validation
of the results.
Keywords: Freight Options; Forward Contracts; Implied Volatility; Economic Modelling; Fundamental
Analysis.
JEL Codes: G13, R40.
1 Int roduction
Freight rates are among th e most volatile asset classes. While the time serie s and
cross-sectional properties of freight rates and t hei r volatility have been investigated
extensively in the liter at u r e, the causes of volati l i ty are less well understood. In
this study, we analyse empirically the drivers of freight market volatility. We use
several macroeconomic and shipping-related factors that are known to a↵ect the
supply and demand for shipping and examine t h ei r impact on the term structure of
freight options implied volatilit i es (IV). The st u d y of option’s IV is a novel area in
the shipping economics and finance literature. This is a forward-looking measure
of vol at i l i ty that is priced i n the ma r ket and reflects the ex pectations of freight
market volatility at the maturity of the corresponding option. At the same tim e ,
it i s a model-free est i m a t e of vola t i l i ty and thus not dependent on t h e specification
or parameterisation of statistical models.Understanding IV better and being able
to forecast it is critical in hedging decisions and in pricing freight options.
Previous studies in the shipping literature used statistical models of volatil-
ity, such as condit i o n a l het er o skedasticity models, which were based on historical
freight rates. See Kavussanos and Nomikos (2000), Lu, Marlow, and Wang (2008),
Chen et a l . (2014) and Dai, Hu, and Zhang (2015). The latter found significant
volatility spillover e↵ects across di↵erent vessel markets and across vessel p r i ces
and freight rates. Kavussano s (1996) examined volatility as a measure of risk in
the dry-bu l k ship market and found that time-charter rates were more volatile than
spot rates and small vessels were less risky than larger ones. Chen, Meersman, and
de Voorde (2010) investigated th e interr ela t i o n sh i p s in daily returns and volati l i t i es
between Capesize and Panamax freight rates in major tradi n g routes a n d found
that the dynamics between the two markets changed across time on di↵erent trad-
ing routes. Al i za d eh and Nomikos (2009) and Tsouknidis (2016) studied dynamic
volatility spillovers using mu l ti variate DCC-GARCH models. These papers would
substantia t e the idea of inter-connectivity between the Capesize and the Panamax
classes but also indicated di↵erences between t h ese classes.
Chen and Wang (20 04) showed a significantly negative relation between returns
and volatility for three di↵erent types of bulk carriers. The e↵ect is stronger in
market downturns than in market upturn s which suggests an inverse relati o n sh i p
between spot rate levels and freight rate volatility, consistent with the notion of a
leverage e↵ect (Black, 1976). Xu, Yip and Marlow (2011) studied the relationship
between freight rate volatility and sup p l y of shipping services and foun d that the
change in fleet size positively a↵ects freight rate volatility, part icul ar l y in the larger
ship classes.
Alizadeh and Nomikos (2011) investigated the relationship between the dy-
namics of the term structure of period rates and time-varying volatili ty of ship-
ping freight rates and found th e r el a t i o n sh i p t o be asymmetric in the sense that
1
when the freight market is in backwardation, volatility is higher compared to pe-
riods when the market is in contango. Aliza d eh (2013) also found that FFA price
changes had a positive impact on trading volume, suggesting a momentum e↵ect
as higher capital gains encourage more transact i o n s. Finally, the importance of
incorporating macro-economic factors in modelling freight rate volatility was also
highlighted by Drobetz, Richter and Wambach (2012).
All of the above studies indica te that freight rate vol a t i l i ty is a↵ected by a
number of idiosyncratic factors as well as factors related to the general state of the
world economy. At the same time, volatility estimates used in those studies are
based on histo r i ca l data and are model-dependent, con d i t i o n a l on the specification
of the statistical model used for their estimation; i t may well be the case that
di↵erent statistical models of volati l i ty will generat e di↵erent results. Since implied
volatilities are forward-looking and model-free estimates of volatility , we overcome
both of those limitations. As such, the proposed framework enables us t o examine
in a robust way how chang es to macro- or shipping-related market conditions a↵ect
the expectations of freight market volatility.
Our aim is to understand the drivers and fundamentals of freight rate volatility
and, in so doing, establish a stronger economic basis in analysing a very useful input
to the pricing and hedging of freight options. We examine a range of supply and
demand-related factors in our models. For supply factors we use the size of the
fleet, orderbook and net cont r a ct i n g . For demand factors, we use variables that
reflect world seaborne trade and world economic activity. In addition, we consider
factors r el a ted to the freight market and the second-hand market for ships such
as, freight market momentum, second-hand sales purchase (S&P) transactions
and second-hand and new-building prices. Finally, we al so consider economy-
wide financial conditions as well as as market conditions in the Forward Freight
Agreements (FFA) market.
We st u d y a number of models in explaining the IV dynamics and it appears
that the most significant predictive variables of monthly IV levels are its lagged
value, spot frei ght rat es , forward FFA curve slope, tradin g volume, the VIX index,
OECD industrial production, China’s industrial production growth, China’s coking
coal imports and ship building new orders. These factors are particularl y r el evant
for the larger class of Capesize ships. For the Panamax class, the various market-,
demand-, and supply-relat ed variables produce the sa m e impact as in the Capesize
class, although their ex p l an at ory power appears to be stronger. Crucially, we find
that im p l i ed volati l i ty is inversely related to the level of spot rates, forms a V-
shaped curve again st the forward rate slope and appears to be directly a↵ected by
the trading volumes in the FFA freight market. The V-shaped ob ser vation is an
interestin g finding - it implies IV increases with c ontango as well as with normal
backwardation. We find IV to increase with supply drivers such as order book
2
or fleet growth; we suggest this could be related to the forward looking negative
impact on spot r at es whi ch induces great er un ce rt a inty for the ship owners and
increases the demand for h ed g i n g . The latter would push up put prices and increase
at-the-money volatility. We find IV to decrease with demand drivers, such as
OECD industrial production and seaborne trade. Higher economic acti v i ty also
appears to reduce IV; the higher certainty of profitability fo r sh i powners app ea rs
to have a calming e↵ect on the hedge market with lower IV. Fina l l y, we find
higher VIX, proxying for higher economic uncert ai nty and investor fear, is related
to a high er IV, though the statistical evidence on the latter is weak, and not as
pervasive as suggested in Robe and Wa l l en (2016) for the crude oil market.
The structure of this paper is as follows. In section 2 we discuss the monthly
data employed in this study. Section 3 contains the empirical results and discussion
of the results. Section 4 provides robustness for our results by considering weekly
data, an expanded uni verse of supply and demand factors and panel regressions.
Finally, Section 5 concludes.
2 Freight and Economics Data
The recent years have been characterized by high volatility in the freight market
and a corresponding growth in the derivatives market for freight. Traditionally,
this market has been used by players in the physical freight market - such as
shipowners, operators an d trading houses - to hedge their freight risks, though this
is now changing rapidly with the increasing par t i cip a t i o n of investment banks and
hedge funds. Market parti ci p a nts trade forward contracts on shipping freight rates,
known as forward freight agreements (FFAs). These are contracts to settle t h e
average spot freight rate over a specified period of time. FFA contracts also serve
as the underlying asset for freight options. Freight options are negotiated over-the-
counter (OTC) and subsequently cleared through a clearing house. The options
market has gai n ed in popularity over the recent years, reaching an equivalent
trading volume of 280 million tonnes of cargo for 2018 and an open interest of 200
million tonnes of cargo, as of December 2018.
Freight options belong to th e wider family of Asian options. In general, Asian
options provide a good defense against market manipulation of the underlying spot
price prior to settlement, since the settlement price of the option is given by the
average of the spot prices over the trading days of the set tl e ment month. Further,
the average value is less exposed to extreme movements at maturity resulting in
option prices which are lower than the prices of - otherwise identical - plain vanilla
options. For these reasons, Asian options are popular in thinly traded or highly
volatile markets, such as the market for freight.
We focus on the Capesize and Panamax sectors as these are the most liquid
3
