PRODUCTIVITY AND EFFICIENCY OF NIGERIA’S SEAPORTS: A Production Function Analysis

PRODUCTIVITY AND EFFICIENCY OF NIGERIA’S SEAPORTS:

A PRODUCTION FUNCTION ANALYSIS

• Productivity and efficiency are key expectation of every production decision.

• The need for efficient utilization of scarce resources calls for a thorough understanding of production processes.

• According to Edwin et al (2002), production processes are concerned with how scarce resources (input) are used to produce a firm’s product or service (output).

• However, production function is defined as the relation showing the maximum amount of output attainable from any specified set of inputs, given the existing technology (Maurice and Thomas, 1999).

• A port’s production function therefore, simply specifies the efficient relationship between its production inputs and outputs.

• While Port productivity is the ratio of what is produced to what is required to produce.

• Efficiency on the other hand, is a measure of how well the observed output differs from the maximum output attainable from the utilization of production resources or inputs.

• Technical efficiency is therefore achieved when the maximum possible amount of output is being produced with a given combination of inputs (Maurice and Thomas, 1999).

• The pressure on ports in the global market place to increase productivity and efficiency arises from two key developments:

• Technological progress in shipping; containerization and growth in ship size and speed. Containerization came with major changes in cargo handling methods using equipment such as cranes, rolling stocks and specialized trucks which accommodate the demands of containerization.

• Ships also grew in size and speed; thus imposing greater demands on port infrastructures and services.

• the fiscal crisis of the 1980s to the mid-1990s in most developing and transition economies (Estache et al, 2002): National government as a result, found it increasingly difficult to continue to finance public monopolies like ports and railways which had hitherto became inefficient.

• Hence, This led to decay of transport facilities and hence poor service delivery in seaports.

• Policy response by most port authorities was the introduction of series of port reforms ranging from deregulation to attract private capital in port financing to port management and administration etc.

• The Nigerian ports authority for example, introduced reforms (in phases) to improve the productivity of the national ports.

• These range from tariff reduction, investment in new infrastructure and superstructure, cabotage legislation to outright port deregulation.

• Despite these reforms, cargo dwell times in the ports are higher than industry standards.

• Port users are offered services that are below their expectations hence the persistence of cargo diversion to neighbouring countries ports, constraints to trade facilitation and negative impact to trade growth.

• The effectiveness of these reforms needs to be assessed to check for efficiency and effectiveness in meeting the desired reform goals.

• It is therefore the objective of this paper to assess the productivity and efficiency of Nigeria’s seaports through the application of a Production frontier analytical model.

• Specifically, the objectives of the study include:

• To find the determinants of Nigerian ports productivity and their elasticities.

• To estimate the total productivity of Nigeria’s seaports.

• To estimate the relative technical efficiency of Nigeria’s seaports.

• Two distinct measures have usually been employed for estimation of port productivity. One method assumes a single output and multiple inputs variables and uses Stochastic Frontier Analytical framework.

• Notable studies using this framework include Aigner and Schmidt (1977), Jondrow et al. (1982), Greene (1980), Stevenson (1980), Lee (1983), Cornwell and Schmidt (1996), Kumbhakar (1990), Battese and Coelli (1988;1992).

• However, the single output measurement method cannot be applied where multiple output measures of a port are considered.

• This is an advantage of the second technique of productivity estimation known as the Data Envelopment Analysis. Examples of such studies include: Farrell (1957) and Charnes and Rhodes (1978). The present study focuses on the application of Stochastic Frontier method on port data from a developing country.

• Data for this study was obtained from NPA statistical records. Although, the Authority has transferred cargo handling and warehousing functions to the private operators, it provides ancillary services, collects and collates data covering cargo handling activities at the seaports.

• Hence, the data for this study was obtained from NPA records on cargo and vessels traffic. It consists of panels on eight coastal ports for traffic activities covering a period of 22 yrs.

• Four variables are considered from the data; they are: Cargo throughputs, Number of berths, Number of Equipments and Stevedoring labour productivity. These were used to model the productivity frontier of the seaports.

• Data Presentation: Descriptive Statistics• Table 1: Descriptive Statistics of Port Metrics

Port Metrics (Mean Values)Ports Thruput1.(mt) Berths Equipmnt Labour(ngh*)

Apapa 4,215.98 22 188 18.348

Cont_Termnl 2,071.28 6 100 18.652

Tincan Islnd 2,571.45 14 140 19.043

Roro 604.799 3 70 17.391

Warri 746.832 22 116 19.478

PHC 1,680.05 8 104 19.087

Onne 1,090.92 5 8 19.174

Calabar 203.643 14 74 17.783

• Table 1 above shows the distribution of port metrics by respective ports for the study period. In terms of throughputs, Apapa port tops the list; totaling over 4 million metric tonnes of non-oil import and export cargoes handled during the study period. Calabar port handled the least quantity of cargo.

• However, the Radar chart below enables a comparison of the ports on all the performance metrics taken together.

Apapa

Cont_Termnl

Tincan Islnd

Roro

Warri

Phc

Onne

Calabar

0

5000

• From the chart, Apapa, Tincan, Container terminal, Port-Harcourt and Onne ports performed well based on all the measured variables in that order of priority.

• Ro-Ro, Warri and Calabar ports however, did not perform well on all the attribute variables.

• The productivity and relative efficiencies of the ports are therefore assessed using the Frontier productivity model discussed as follows.

Assuming that there are k- inputs and that the production function is linear in logs, and defining ; the production function becomes:

The terms are assumed to be distributed independently of each other and the covariates in the model.

01

) ln( )ln( ... (1)k

j jit it itj

it x V Uy

ln( )it itU

Vit itU and

Based on the forgoing, the research model for this study is therefore specified as:

where • Thruput: cargo throughput (dependent Variable)• Berths: Number of berths in a port• Equipmnt: Number of available equipment• Labour: stevedoring labour productivity (NGH)• And is the decomposed error term.• Results of statistical analysis is presented below.

0 1 2 3ln( ) ln ln ln ... (2)it it it itthruput Berths Equipmnt Labour

it

Table 2: Production Frontier Analysis Results.

Coef. Est. Std. Err. t P>|t| Est. Std. Err. t P>|t|

 1.076 0.234 4.600 0.000 1.028 0.152 6.760 0.000

  -0.184 0.071 -2.610 0.009 -0.169 0.072 -2.340 0.019

  -0.203 0.169 -1.200 0.229 -0.212 0.167 -1.270 0.204

_Cons 12.581 0.878 14.330 0.000 13.588 0.639 21.260 0.000 

- 19.354 179.964

  - 0.982 0.165  

- 19.01 179.965

  - 0.344 0.039 Log L -161.532 -163.987

Berths

Equipmnt

Labor

2

2u

2v

Findings from the Results

• In table 2, two frontier model are estimated: Generalized least square and a time invariant model based on truncated normal distribution of error term.

• In both models, all the production inputs except labour are significant at .

• The coefficient of berths is positive; indicating that available berths at the port contributes to its productivity.

• Equipment variable is also significant but has a negative sign. The sign may represent additional labour (over-booking) when marginal product of Labour is at point of Diminishing Return.

= 0.05

Table 3: Marginal Effects & Productivity of Physical Products

Variable Std. Err. t P>|t|

Berths 1.028 0.152 6.760 0.000

Equipmnt -0.169 0.072 -2.340 0.019Marginal effects after xtfrontier

y = Linear prediction

y =2,473,134.667

/y x

FINDINGS FROM THE RESULTS

• In table 3 above, marginal values of the marginal physical product of capital inputs are 1.028 and -0.169. These represent the proportions in which the inputs can be combined to produce total factor productivity.

• As shown in footer of the table, the total productivity is 2.5 million metric tonnes per annum.

Findings from the Results

• In figure 1, relative efficiency values are calculated for each port. Results indicate that Apapa, Tincan, Ro-Ro, PHC and Onne ports have technical efficiency values of approximately 1.0.

• However, Container terminal, Calabar and Warri ports operate at efficiency levels 1.0 and hence are considered technically inefficient.

Findings from the Results

.081419

.567879

.847193

.17879

.882826

.875445

.325268

.751995

0 .2 .4 .6 .8Technical Efficiency

Calabar

Onne

PHC

Warri

RORO

TCIP

Cont_Terminal

APAPA

Efficiency Analysis of Nigeria's Seaports

DISCUSSION OF RESULTS

• The major finding of the study are: total factor productivity of Nig. Seaports averages 2.5 million tonnes/annum. Number of available berths and cargo handling equipments significantly contribute to port productivity. Labour input is not significant suggesting the use of low productive dock labour in cargo handling operation.

• Among the ports studied, only Container terminal, Warri and Calabar ports were found technically inefficient.

• The major problem is under utilization of infrastructure caused by inaccessibility to these terminals. Calabar port lacks good rail/road connections. Eastern port users prefer other ports in the same range. Warri has draught restriction and does not accommodate large cargo vessels. poor port patronage lead to low throughputs handled and hence under utilization of cargo handling infrastructure.

CONCLUSION

• Low efficiency recorded in these ports for the most part, are due to under utilization of port infrastructure. The federal government should improve road/rail connection to Calabar port and dredge Warri port. Again, there is need to mechanize cargo handling operation to boost labour productivity in the seaports.

• Future studies should compare empirical total productivity value of Nig. Ports with global industry standards to establish optimality or otherwise.

• A hybrid production frontier model should be developed that incorporates the advantages of DEA and SFA models.