WMU

1
WMU
An estimation of the systematic risk of monthly
Capesize spot and time charter rates (1999-2004)
Pierre CARIOU - World Maritime University
(Sweden)
IAME 2006 - Melbourne
2
Background
WMU
Capesize is the name given to dry bulk carriers
larger than 80.000 or 120.000 deadweight tons
according to classifications. In order to be
profitable two conditions have to be fulfilled
large shipments and long distances. Two main
commodities are able to offer these two
conditions iron ore (68 of Capesize capacity in
2002) and coal (31).
3
Objective
WMU
Higher volatility in prices (risk) compare to
other dry bulk vessels (Handymax, Panamax...)
leads to higher returns. Is it true for every
route/contract? Using CAPM, the return for a
particular contract could be explained by the
behaviour of price compare to other spot and time
charter rates.
4
General findings from literature
review Theoretical foundations
WMU
In the short run the relative inelasticity of
demand and supply lead to the volatility in
prices (Koopmans 1939, Zanetos 1966, Beenstock et
al. 1993, Stopford 1997). In the long run
Scrapping and/or delivery of vessels drive the
freight rates up/down and lead to the cyclical
behaviour of prices (Adland et al. 2004).
5
General findings from literature review Empirical
investigations on Capesize
WMU
Higher volatility in price (Veenstra, 1999,
Adland et al., 2004, Kavussanos, 1996, 1997,
2002, Alizadeh et al., 2003, Tsolakis et al.
2003). Higher seasonal fluctuations (Kavussanos
et al., 2001) Market conditions (in a booming
period, the demand crosses the inelastic part of
the supply curve) and contract duration play on
volatility (Kavussanos 2002, Chen et al., 2004).

6
The methodology
WMU

• To our knowledge, only a few articles have
  applied Portfolio theory to shipping
• Kavussanos et al. 2003 used CAPM for shipping
  industries
• Cullinane (1995) used the Markowitz portfolio
  theory to analyse the dry bulk freight market.

7
The methodology
WMU
The estimation of the systematic risk or beta
value is usually done through OLS. Rit is the
return for asset i in period t, Rft, the
risk-free rate, RMt, the return on the market
portfolio in t and eit the residual or
non-specific risk. ?i and ?i are the CAPM
parameters for stock i.
8
The methodology
WMU
A negative/positive ?i indicates that the stock
is over/under priced and ?i value greater than
one carries above average risk and should lead to
higher returns. We assume that the return that
can be expected from a specific contract can be
approached by the spot and the time charter rates
in index (change in cost index is similar) so
that
9
The sample
WMU
10
The sample
WMU
11
Statistical properties of the sample General
indexes
WMU
12 monthly prices into logarithm from January
1999 to December 2004 for contracts on Capesize
vessels carrying iron ore (Drewry Monthly
Statistics, SSY, Fearnleys Monthly, the Baltic
Exchange and ISL). For general indexes (table
2), the mean of our 12 routes (2.0116 for i-12)
is higher than the mean of BPI, BCI and BDI.
Prices are not stationary in level but
stationary in first difference at conventional
levels.
12
Statistical properties of the sample General
indexes
WMU
Jan1999-Dec2001, the mean and s.d. of Capesize
index (BCI and i-12) are below BDI BPI and then
above. Three main reasons can be stressed.
Firstly, the routes we selected are not similar
in the BCI, BPI and BDI (see Baltic Exchange for
definition). Secondly, routes are equally
weighted (1/12) for i-12, while they are not for
the BCI, BPI or BDI.
13
Statistical properties of the sample Individual
routes
WMU
Thirdly, a shortage in capacity that is the
consequence of the increase in Chinese iron ore
imports in tons but also in tonne-miles. EU-15
and Japan still generate around 57 of the
Capesize demand in tonne-mile in 2003, while
China only 13 but the market could be in a
situation where the demand crosses the inelastic
part of supply.
14
Statistical properties of the sample Individual
routes
WMU
The price with the highest mean (trip charter
rates - 7 to 10) has the highest s.d. For
spot rates (1 to 6), route 4 (W. Austr. - China)
has the highest return mean (2.0134) but not the
highest s.d. (W. Austr. - ARA). For trip
charter rates, route 7 (Cont/Transatlantic rv
120,000 -16,000 dwt - 5-15 yrs) has the highest
return mean (2.016) but not the highest s.d.
(route 9 FE/-/Cont - 140-170.000 dwt - 1-10 yrs).
15
Beta value
WMU
The return/systematic risk relationship is
determined by the possibility to spread the risk.
In our specific case, systematic risk is
apprehended comparing individual routes with
indexes such as i-12, BCI, BPI or BDI.
16
Beta value
WMU
17
Beta value
WMU
Low Beta value return Spot and T/C rates
High Beta value return Trip charter rates
Undetermined
18
Empirical results
WMU
On the sub-period Jan 2002 - Dec 2004, a general
increase in the beta-value and in the return for
every contract as expected with the notable
exception of contract 10 (Sing-jap/Australia rv -
140-170.000dwt - 1-10 yrs ). The BCI (beta
value of 1.11) over-reacted compared to BDI. In
line with previous research for which Capesize
vessels over-react in booming periods.
19
Implications
WMU
The trip charter routes that represent around 40
of the weights of the BCI in 1999 (route C8 to
C11 in Baltic exchange) are the route with the
highest beta value. In April 2004, a new trip
charter route was also added (C12 - Gladstone to
Rotterdam 150,000 dwt) that pushed the weight
of trip charter routes to 50.
20
Conclusion
WMU
The estimations do not show an over-reaction for
all contracts but mainly for trip charter rates.
This last finding suggests that the composition
of indexes and the weight given to trip charter
rates in the BCI are not neutral to understanding
the general behaviour of price indexes and
especially in booming period.
21
Conclusion
WMU
Thank you for your attention QA Session