Measuring the market risk of freight rates: A forecast combination approach

Published date01 March 2018
DOIhttp://doi.org/10.1002/for.2485
Date01 March 2018
Received: 25 January 2016 Revised: 11 March 2017 Accepted: 26 May 2017
DOI: 10.1002/for.2485
RESEARCH ARTICLE
Measuring the market risk of freight rates: A forecast
combination approach
Christos Argyropoulos Ekaterini Panopoulou
Kent Business School, University of Kent,
Canterbury, UK
Correspondence
Ekaterini Panopoulou, Kent Business
School, University of Kent, Canterbury CT2
7FS, UK.
Email: a.panopoulou@kent.ac.uk
Abstract
This paper addresses the issue of freight rate risk measurement via value at risk (VaR)
and forecast combination methodologies while focusing on detailed performance
evaluation. We contribute to the literature in three ways: First, we reevaluate the
performance of popular VaR estimation methods on freight rates amid the adverse
economic consequences of the recent financial and sovereign debt crisis. Second, we
provide a detailed and extensive backtesting and evaluation methodology. Last, we
propose a forecast combination approach for estimating VaR. Our findings suggest
that our combination methods produce more accurate estimates for all the sectors
under scrutiny, while in some cases they may be viewed as conservative since they
tend to overestimate nominal VaR.
KEYWORDS
backtesting, combination forecasts, freight rates, performance evaluation, value-at-risk, volatility
forecasts
1INTRODUCTION
Freight rate risk has always been one of the most impor-
tant risk factors of the shipping industry, mainly because it
affects its primary source of income. The rising interest of
participants in the shipping industry such as shipping com-
panies, shipping hedge funds, and shipping banks makes the
accurate measurement of freight risk a procedure of high
importance and difficulty induced by the intrinsic characteris-
tics of the shipping industry itself. Specifically, the cyclicality
of the maritime economy and the mechanics of the ship-
ping markets create a complex profile for both the freight
rates and their volatility.1While,from a financial perspective,
freight rates are typically considered part of the commodi-
ties market, freight rate markets are quite different from the
majority of the other commodities markets. For example,
in contrast to all major traded commodities, freight rates
are essentially not storable—a property that makes simple
cost-of-carry valuations of futures contracts for freight rates
1For an in-depth discussion on freight rate characteristics, see Stopford
(2009) and Alizadeh and Nomikos (2009).
impossible. Moreover, the freight rate spot market shows a
high degree of volatility and seasonality, which causes signif-
icant risks for shipowners, charterers, and market participants
in general (Alizadeh & Nomikos, 2011 Alizadeh, Huang, &
Dellen, 2014). Consequently, producing accurate estimates
of freight risk is essential to freight market participants as it
enables them to enhance their ability of sound strategic invest-
ment and hedging decisions. This paper addresses the subject
of freight rate risk measurement via Value at Risk (VaR)
and forecast combination methodologies while focusing on
detailed performance evaluation.
There are many factors behind the fluctuation of freight
rates. In the long run, freight rates are determined through
the interaction between the supply-and-demand schedules for
shipping services. Information on vessel availability and pro-
duction directly influences price levels. On the other hand,
demand for shipping services is closely linked to the busi-
ness cycle and economic growth through various channels.
For example, during economic booms, both production of
commodities and demand for crude oil is high. This leads
to increases in both dry bulk (responsible for raw nonliq-
uid materials transportation) and tanker (responsible for the
Journal of Forecasting.2018;37:201–224. wileyonlinelibrary.com/journal/for Copyright © 2017 John Wiley & Sons, Ltd. 201
202 ARGYROPOULOSAND PANOPOULOU
transportation of liquids, such as crude oil and petroleum
products) freight rates. In the short run, the cost of oper-
ating a vessel greatly fluctuates with the cost of crude oil,
mainly used as a shipping fuel and known as “bunker.” Con-
sequently, bunker prices are closely linkedto cr ude oil prices
and, therefore, freight rates react to changes in oil price lev-
els. Moreover, tanker vessels account for approximately half
of the world's seaborne trade, and tanker freight rates deter-
mine the transportation cost of strategic for world economy
products such as crude oil and its by-products. Finally, the
importance of freight rates for global real economic activity is
highlighted in the index of real economic activity constructed
by Kilian (2009). This index is essentially based on global
dry cargo freight rates and exhibits a high correlation with
the Baltic Dry Index (BDI). BDI is often employed as a gen-
eral market indicator or “barometer”, reflecting changes not
only in the dry-bulk market, but also in the overall worldwide
real economy.
Given the importance of freight rates, calculating freight
rate risk accurately is of utmost importance for at least
three reasons. First, market participants can develop hedg-
ing schemes more effectively and efficiently when they are
aware of the risk they are exposed to. Simple risk metrics, like
volatility, have not been proven adequate for this market due
to deviation from normality, complexity, cyclicality, and the
existence of jumps during extreme events (see Kavussanos &
Dimitrakopoulos, 2011; Kavussanos & Visvikis, 2004). For
example, Alizadeh and Nomikos (2011) argue that volatility
dynamics vary with shipping market conditions; that is, they
are regime dependent. In the same vein, Alizadeh et al. (2014)
implement a regime-switching multivariate approach in order
to capture the volatility dynamics and possible correlations of
spot and futures prices in the tanker sector. Second, during the
last decade, the shipping freight market has transformed from
a service market, where freight rate was the cost of transport-
ing raw materials by sea, to a market, where freight rate is
seen as an investment like any other assetor commodity (see,
e.g., Nomikos, Kyriakou, Papapostolou, & Pouliasis, 2013).
Market participants now include investment banks and hedge
funds, and are interested in quantifying the risk profile of
this alternative asset class having realized its potential bene-
fit for both speculation and diversification. Alizadeh (2013)
documents a positive contemporaneous relationship between
trading volume, which has increased rapidly in recent years,
and volatility in the shipping forward freight market. Finally,
VaR provides a mean of setting margin requirements in the
freight exchange derivatives market, which is expanding fast.
Given that forward freight agreements (FFAs) and freight
options are employed to hedge freight rate risk and that trad-
ing of these derivatives can be done through an organized
exchange, margin determination is very important. With the
elimination of credit risk, margins reflect market liquidity and
volatility of the underlying spot freight rates.2Specifically,
FFAs are the primary instrument shipping market participants
employ to hedge freight exposure risk. These contracts are
agreements between a buyer and seller to settle the difference
between the contract price and an appropriate settlement price
in cash. The settlement price is normally the average of the
spot freight rates on the underlying shipping route over the
calendar month, reported by the Baltic Exchange. Alternative
freight risk management techniques include time charter con-
tracts, contracts of affreightment (CoAs), and freight options.
While period charter contracts and CoAs are considered phys-
ical forms of hedging, these contracts are not very liquid and
operationally flexible. Finally,freight rate options can be used
for hedging freight rates, but they are not very liquid and are
comparatively expensive.
Despite the plethora of related research in the financial
sector, measurement of the market risk of freight rates has
been under-researched. To the best of our knowledgethe most
recent contributions in the field of freight rate risk adopt the
VaR methodology in order to measure the market risk of the
dry bulk and tanker freight market. Specifically, Angelidis
and Skiadopoulos (2008) applied a variety of parametric, non-
parametric and hybrid methods in order to measure the market
risk mainly for the dry bulk sector. Their findings suggest
that in almost all cases the simplest nonparametric meth-
ods produce accurate results. On the other hand, Kavussanos
and Dimitrakopoulos (2011) dealt with selection of the appro-
priate freight rate risk model by applying a similar VaR
methodology solely for the tanker sector. The authors find
that parametric methods are more suitable for this sector.
More recently, Abouarghoub etal. (2014) utilized a two-state
Markov switching distinctive conditional variance model in
order to improve the tanker sector VaRforecasts. Their results
suggest that a regime-switching approach can capture more
precisely the tanker sector volatility dynamics, thus providing
better VaR forecasts.
In this paper, we contribute to the literature on both tanker
and dry bulk freight rate risk forecasting in the following
dimensions. First, we reevaluate the performance of pop-
ular VaR estimation methods amid the adverse economic
consequences of the recent financial and sovereign debt cri-
sis. Second, we provide a detailed and extensive backtest-
2Predictability in the underlying freight spot rate does not imply predictabil-
ity of the corresponding derivative contract, since the standard cost-of-carry
relationship for financial futures does not hold for the freight ones. This
is because the underlying asset is not tradable, and hence the pricing by
arbitrage argument cannot be applied. A series of papers have tested the
unbiasedness hypothesis of the market, that is, whether the freight rate mar-
ket is efficient. Results are mixed, mainly due to market segmentation (see
Kavussanos & Visvikis, 2006 , for a review).A more recent contribution, by
Goulas and Skiadopoulos (2012), points to the inefficiency of the Interna-
tional Maritime Exchange (IMAREX) freight futures markets overt he daily
horizon. Futures trading strategies based on the formed daily forecasts the
authors develop yield a positive, economically significant risk premium.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT