Granger Causality Test Stability Analysis

33 Wood Product r LR p-value Number of Lags Logwood 24.80 0.0659 2 1 5.92 0.4813 Plywood 34.99 0.0021 2 1 7.65 0.2905 Sawn wood 42.43 0.0001 1 1 4.47 0.6771

5.3 Granger Causality Test

Granger causality test is conducted to indicate the presence of price transmission between two markets and to which direction. As shown in the table 6, the result of Granger-causality test does not find any Granger-causality for logwood as both the null hypothesis are not rejected. Meanwhile, for plywood and sawn wood, the Granger-causality test produces the same result i.e. reciprocal causality in which both the null hypothesis to the extent that domestic prices do not affect world prices and that world prices do not affect domestic prices are rejected at five percent level of significance. According to granger causality test, thus it can be concluded that the domestic price and the world price of plywood and sawn wood have an influence to each other, and on the contrary, the domestic and the world price of logwood have no influence each other. Table 6 The Result of Granger-Causality Test Wood Products Null Hypothesis Ho Test- Statistics P-value Logwood  “dl” do not Granger-cause “rwl” 1.4600 0.2338 34  “nwp” do not Granger-cause “dp” 0.3058 0.7368 Plywood  “dp” do not Granger-cause “rwp” 8.1551 0.0004  “rwp” do not Granger-cause “dp” 3.3617 0.0359 Sawn wood  “ds” do not Granger-cause “rws” 6.9326 0.0011  “rws” do not Granger-cause “ds” 3.1179 0.0456

5.4 Stability Analysis

Stability analysis is conducted to evaluate the parameter instability in the VECM due to the presence of structural break. If we observe the sample split figures, it is found that only in the plywood model, all the bootstrap p-values are much higher than 0.1, which implies that all the parameters in the model are stable. Meanwhile, in the logwood model, there is one break point in which the bootstrap p-value is lower than 0.1. Furthermore, in the sawn wood model, generally it shows the evidence of instable parameter in the model, in which almost half of the bootstrap p-values are less than 0.1 as well as 0.005. Then, if we observe the break point figures, it is showed that all of the models are not stable in the parameter which is indicated by the presence of many break points. The presence of break point either in the sample split chow-test or in the break point chow-test could indicate the presence of structural change in the wood industry. Based on those results, it is indicated several structural changes in 2006, 2008, and 2009. The presence of structural change in the sawn wood model in 2006 might be related the restriction of sawn wood imported from Indonesia by United Kingdom and several European countries. Consequently, the export of Indonesian sawn wood declined substantially, which was 35 percent less than 2005 levels. At the same time, a structural change might also occur in Malaysian market, due to the increase diversion of logwood to plywood mills and thus depressed the supply of logwood for sawn wood mills. In addition, the economic crisis particularly in Europe and USA which started from 2008 could also have implication on the structural change on the wood industry. Furthermore, in 2009, 35 Indonesian government amended the log export ban policy, which allows for export of several types of logwood. Logwood Sample Split Break Point Plywood Sample split Break Point Sawn wood Sample Split Break Point Figure 8 The Results of Chow Test

5.5 Model Selection