Results and discussions

Based on the models for measuring “return to the dollar”, technical efficiency and allocative efficiency shown in equation (5.9), the changes at the station-level of TMTC before and after privatization were evaluated according to the pre-selected indicators described in the previous section. Then technical efficiency and allocative efficiency are measured, and can be gauged further the extent of distortions created in the bus service market. Lastly, the origin of “return to the dollar” is tried to be identified; that is, from TE or AE?

Looking first at Table 5.2, there is a considerable discrepancy between profit margins before and after privatization. Despite the privatized firm having an average profit margin

Table 5.2 Decomposition of “Return to Dollar”

TR/TC Technical efficiency Allocative efficiency Station

Rating Rank Rating Rank Rating Rank TMTC

Total geometric mean 0.918 0.880 1.041 Test of

significance p-value 0.017* p-value 0.044* p-value 0.130

Note: (1) TR/TC denotes “total revenue” divided by “total cost”.

(2) Paired difference experiments are used to test for the same mean between two groups.

(3) “*” means significant at the 5% level of significance.

of about 4.1%, the former public firm incurred losses with total costs exceeding total revenues by 19% on average. In contrast to only 3 out of 15 stations that made profits before privatization, 10 out of 15 stations operated at a profit following privatization. A ranking of pre-privatization performance is very different from a ranking of post-privatization performance across the whole sample of stations. With regard to the decomposition of the “return to the dollar”, the results show that the two firms as a whole suffered from low levels of efficiency (88%), and the SOE with a technical efficiency score averaging 84.2% lagged behind the POE, which had a technical efficiency average of 92.3%.

However, as the figures in the allocative efficiency columns indicate, perhaps in an attempt to cover the inefficiency-induced losses, both the SOE and POE adopted distorting relative output prices with respect to inputs prices as an expedient; with the distorting being more pronounced in the POE than in the SOE.

The non-parametric technique reveals that both components of “return to the dollar”, TE and AE, increased immediately following privatization. The decomposition of the

“return to the dollar” helps to guide the search for an explanation for the measured profit margin change. In this case, the decomposition indicates that this increase in profit margin appeared to be mostly attributed to allocative progress rather than to improvement of technical efficiency. This will be discussed later. However, the statistical tests show a statistically significant increase both in profit margin, with a p-value of 0.017, and TE with a p-value of 0.044, while there is an insignificant increase in AE with a p-value of 0.130.

The average efficiency scores associated with each of the three station types, i.e., northern (N) region, central (C) region, and southern (S) region before and after privatization are shown in Table 5.3, along with averages for the entire group of 15 stations.

More specific origins of these performance changes can be seen in the following station type breakdowns.

Table 5.3 Decomposition of “Return to Dollar” by station type

Station type TR/TC Technical Allocative

TMTC (2000)

N 0.870 0.852 1.021

C 0.805 0.835 0.964

S 0.734 0.833 0.882

All 0.810 0.842 0.961 GGBC

(2002)

N 1.139 0.957 1.190

C 1.129 0.982 1.150

S 0.874 0.844 1.035

All 1.041 0.923 1.128 Change

(2000-2002)

N 1.309 1.123 1.166

C 1.402 1.176 1.193

S 1.191 1.013 1.173

All 1.285 1.096 1.174

As can be noted in the upper part of Table 5.3, the stations of the northern region dominated the other station types in all three efficiency dimensions. Before privatization, for example, the stations of the northern region incurred losses with total costs exceeding total revenues by 13.0% on the average, less than the 14.8% technical inefficiency, and with an allocative efficiency of 1.021. This may imply the ability to distort output prices at a higher rate (2.1%) than input prices. In the central region group, on the other hand, the average profit margin was about negative 19.5%, with technical and allocative components of 0.835 and 0.964, respectively. As for the stations of the southern region, they averaged 26.6% negative profit margins, a less than 16.7% technical inefficiency, and a less than 11.8% allocative inefficiency.

The post-privatization patterns of performance, as shown in the middle part of Table 5.3, are similar to those of pre-privatization, though with relatively higher performance scores. The stations of the northern region, for example, exhibit a significant increase in profit margin, due to both an increase in the technical and allocative components.

Noteworthy is that the technical efficiency goes from 0.852 to 0.957 between the two organization types. This implies that the stations of the northern region convert their input resources into output more efficiently following privatization. As for allocative efficiency which goes from 1.021 to 1.190, this may imply the ability to distort output prices at a higher rate (16.9 %) than input prices.

The stations of the central region experienced a similar increase in profit margin, due more to an allocative component than to a technical component, and this former figure rose from 0.964 to 1.150 between the years, while the latter also increased from 0.835 to 0.982.

The southern region group also displays a similar pattern, however, as the profit margin change indicates that KKTC still incurred a loss of 12.6%. Consequently, one might say that although the stations of the southern region, in terms of converting input resources into output, made correct decisions after privatization, they were still operating less well than the other two regions’ units, due probably to the competitive pressure in this region being relatively small. Furthermore, the change in AE was bigger than TE for each station type, as portrayed in the lower part of Table 5.3, thereby confirming that the increase in profit margin came mostly from progress in AE rather than from improvement in TE.

There is another interesting phenomenon relating to the scale of individual bus stations.

Figures 5.2, 5.3, and 5.4 depict the inverted U-shaped relationship between the “return to the dollar”, technical efficiency, and allocative efficiency for all the sample stations against fleet size, respectively. As can be observed, the results of the three measures suggest that stations of approximately 40-70 vehicles are of optimal size. This is consistent with Cho and Fan’s (2003) finding that these stations satisfy constant returns to scale of inputs and output.

In summary, depending on the results of Tables 5.3, and previous discussions, it is found that both TE and AE contribute to the growth of profit margin, with AE playing a more important role than TE.

Figure 5.2 TR/TC rating versus number of vehicles (KKTC, 2002)

Figure 5.3 Technical efficiency rating versus number of vehicles (KKTC, 2002)

Figure 5.4 Allocative efficiency rating versus number of vehicles (KKTC, 2002)

S14●