BEMMFISH Conference Barcelona Spain 18 22 October 2004

BEMMFISH Conference, Barcelona, Spain, 18 -22 October 2004 PRODUCTION FACTOR’S ELASTICITIES. EFFECTS OF VARIATIONS OF FISHING CAPACITY AND ACTIVITY IREPA Onlus - Istituto Ricerche Economiche Pesca e Acquacoltura Via S. Leonardo, Trav. Migliaro, Salerno, ITALY http: //www. irepa. org irepa@irepa. org IREPA Onlus

Research project: Bio-economic modelling of Mediterranean Fisheries • general and flexible bio-economic model, which accommodates the realities of the most common Mediterranean fisheries • The model includes multiple species, multiple fleets and gear types • It also implements all fisheries management tools currently used by Mediterranean fisheries managers and a wide range of fisheries performance measures. IREPA Onlus

Objectives • Estimate the short run elasticities of the factors in the Northern and Central small pelagic fishery (GSA 17) • Analysis of long run effects of landings, varying capacity and activity • Estimate the interrelations existing between the resource dynamics and the evolution of landings, capacity and activity IREPA Onlus

The case study Northern and Central pelagic fleet: Italian MAGP 4 H 3 & 4 H 4 • By back programs resulted in a radical reduction of the fleet • Fishing activity is strongly controlled: • Since 1988 a temporary withdrawal has been imposed for about 45 days, between June and September • The same regulation prohibits the activity during the 8 weekends following the temporary withdrawal • The target species, anchovies and pilchards, are generally subject to strong biomass fluctuations, which seem to be linked to the relatively high mortality rate of juveniles, larvae and ichthyoplankton IREPA Onlus

Number of vessels, capacity, activity and landings of N&C Adriatic pelagic fishery (1972 -2000) IREPA Onlus

The VAR and IRA Methodology The Vector Auto Regression (VAR) is a “a-theoretical“ model, commonly used forecasting systems of interrelated time series and for analysing the dynamic impact of random disturbances on the system of variables. It is useful if the structural relation existing between variables are not prior defined. The scenario considered in our model is that evolution of the landings depends not only upon the current and past combinations of landings, capacity and activity, but also upon current and past interrelationships between their mutual effects. Another important output of VAR Analysis is the of Impulse Response function (IRA), which traces the time path or impulse responses for each of the three modelled variables to unitary shock. In the context of our analysis, we assumed that shocks or innovations are variations in capacity or activity planned in the management system IREPA Onlus

The mathematical form of a VAR Where: yt is a k vector of endogenous variables, xt is a d vector of exogenous variables, A and β are matrices of coefficients to be estimated, εt is a vector of innovations. The VAR model simply consists in regressing each current variable on all other variables lagged a certain number of times. IREPA Onlus

The VAR estimated In the context of our model, we know that changes in capacity (GRT) and activity (DAYS) surely affects the output (Y). However, it is possible that variations in landings must also cause change in activity or capacity in the next periods. Similarly, a capacity variation probably causes activity to change and vice versa. In other words, since these variables interacts and are jointly dependent, there is the possibility that they are endogenous. The system of equations estimated are: IREPA Onlus

A preliminary analysis: the Johansen Cointegration test In the trace test (Qr): T is the number of observations; r traces cointegration from zero to k-1; k is the number of endogenous variables; λi is the ith largest eigenvalue obtained from the matrix of estimated parameters IREPA Onlus

The VAR results IREPA Onlus

Dynamic behaviour: The Impulse Response Analysis IREPA Onlus

Conclusions The VAR regression estimated the short run interaction existing between landings, capacity and activity. As expected , the elasticity of capacity at time t-1 vis-à-vis landings (at time t) is positive and is equal to 0. 50. An increase of capacity leads to an increase in landings, not only at time t, but also at time t+1 and so on. In simply words, capacity is a structural factor that produces positive effects on catchability. A negative relation (-1. 35) between fishing days and landings was found. This result can be related to the dynamic of the biomass: an increase of fishing days, affecting the level of stock abundance, decreases the landings in the following year. Finally, it was found a statistically significant and positive elasticity between capacity and activity: a variation of 10% in capacity determinates an increase of 13% in fishing days. IREPA Onlus

Conclusions The results of the Impulse Response Analysis are consistent with those obtained in the VAR regression Through impulse-response functions, it is determined how could be the direction and amplitude of the effects on future landings by varying fishing capacity and/or activity The result indicates that landings react to variations on activity more than to variations on capacity. In fact, the landings impulse responses to variation of activity are very high in the short run period. On contrary, a unitary shock of capacity has a lower but more persistent effect on landings. These results suggest different management approaches based on the stock exploitation level. Reducing fishing effort by varying activity is suggested for stocks overexploited, which need immediate intervention. Reductions of capacity are suggested for stocks normal or under-exploited. IREPA Onlus