wordpress-themes.org wordpress themes wordpress themes

A Meta-Analysis of the Price Elasticity of Meat: Evidence of Regional Differences - Estimation ResultsThe OLS and random effects results for each region are provided in Table 1. As indicated at the bottom of the table, the LaGrange Multiplier test favors using random effects over OLS in each region. Rather than discuss the results of each estimation separately, we will compare the general pattern of the coefficients across the three regions. Concerning the different meats, relative to the baseline meat composite there are notable differences across the three regions. Consistent with Gallet there is a tendency in North America for the price elasticity to be more (less) elastic for beef, lamb, and fish (poultry), whereas the insignificant coefficient of pork implies the price elasticity of pork is statistically in the neighborhood of the meat composite. Yet for the other regions there are fewer significant differences in the price elasticity across meats. Although the signs of the coefficients are similar for North America and Asia, each of the meat product coefficients is insignificantly different from zero in the OLS results for Asia, while the random effects results for Asia indicate significantly more elastic price elasticities only for beef and lamb. For Europe, although beef and pork are more elastic (yet insignificantly so in the random effects results), the price elasticity of fish is significantly less elastic, which runs counter to North America.
Turning to specification issues, the functional form coefficients are most often negative and significant for Europe. Relative to the baseline linear specification of meat demand, other functional forms thus tend to generate more elastic estimates of the price elasticity in Europe. Yet for North America and Asia (especially in the random effects results) the larger share of coefficients being insignificant implies functional form plays a more modest role in
determining the price elasticity. Region differences in the signs of many of the functional form coefficients, notably those that are significantly different zero (i.e., semi-log, AIDS-nonlinear, and CBS), as well as those associated with other specification issues (i.e., substitute meats and two-step treatments), also highlight that demand specification has different impacts on the price elasticity across regions.
As for the remaining issues, the results are mixed. For instance, a greater share of the data-oriented coefficients in the OLS results for Europe are significantly different from zero, compared to North America and Asia. Yet when we examine the random effects results the lack of significance of most of these coefficients suggests data plays a less important role in determining the price elasticity. Although the OLS results for North America suggest the use of 3SLS, FIML, and GLS (GMM) contributes to more (less) elastic estimates of the price elasticity, significance of the estimation method coefficients drops off sharply for Asia and Europe, as well as for the random effects results. Finally, although more elastic estimates of the price elasticity in North America tend to be published in top 36 economics journals or the AJAE, for Asia and Europe the quality of the publication outlet matters less.

Table 1. Meta-regression results

Variable OLS Random Effects
North America Asia Europe North America Asia Europe
Product:
Beef -0.120*(1.898) -0.133(1.597) -0.150*(1.793) -0.154***(3.012) -0.264***(3.021) -0.106(1.359)
Pork 0.052(1.333) 0.040(0.488) -0.104*(1.873) -0.018(0.464) -0.069(0.748) -0.049(0.801)
Lamb -0.316**(2.038) -0.143(1.336) -0.116(0.980) -0.390**(2.224) -0.285**(2.385) 0.006(0.053)
Poultry 0.221***(5.487) 0.004(0.050) -0.020(0.278) 0.156***(4.248) -0.107(1.239) 0.040(0.600)
Fish -0.286***(3.777) -0.054(0.604) 0.076*(1.691) -0.212***(3.996) -0.093(0.909) 0.108***(2.862)
Functional Form:
Double-Log -0.167*(1.877) -0.216(1.310) -0.513***(2.600) -0.092(0.888) -0.180(0.862) -0.511(1.401)
Semi-Log 0.456**(2.299) -0.229**(2.089) -2.057***(5.825) 0.152(0.669) -0.155(1.108) -1.029**(2.387)
AIDS-Nonlinear 0.220**(2.356) -0.072(0.566) -0.550***(5.369) 0.128(1.028) -0.087(0.525) -0.537***(2.904)
AIDS-Linear 0.028(0.361) -0.241**(2.150) -0.530***(5.484) 0.015(0.151) -0.167(1.101) -0.388**(1.991)
AIDS-Quadratic -0.120(0.573) -0.290**(2.509) -0.607**(2.400) -0.207(0.710) -0.343*(1.918) -0.458(1.102)
AIDS-General -0.003(0.035) 0.281**(2.484) -0.021(0.161) 0.208(1.293)
Rotterdam -0.143(1.337) -0.153(1.215) -0.781***(4.083) -0.038(0.322) -0.108(0.361) -0.631***(2.918)
CBS 0.649**(2.573) 0.026(0.168) -0.577***(4.791) 0.403(1.328) 0.119(0.439) -0.454*(1.940)
Translog -0.049(0.765) -0.038(0.380) -1.819***(3.639) -0.094(0.891) -0.113(0.616) -1.772(1.395)
S-Branch -0.357***(4.482) -0.181(0.874)
Box-Cox -0.061(0.679) -0.762***(5.339) -0.068(0.593) -0.439**(2.284)
Other Form 0.090(0.948) -0.422***(2.906) -0.960***(7.010) 0.161(1.285) -0.478**(2.541) -0.638***(5.400)
Variable OLS Random Effects
North America Asia Europe North America Asia Europe
Other Issues:
Compensated 0.067(1.182) 0.219***(4.497) 0.221***(6.737) 0.105*(1.813) 0.186***(3.893) 0.223***(6.135)
Substitute Meats -0.187**(2.571) 0.059(0.981) 0.154***(3.011) -0.185**(2.155) 0.078(0.668) 0.096(1.449)
Two-Step 0.244***(3.029) -0.158**(2.397) -0.154(1.006) 0.197*(1.696) -0.112(1.161) -0.254(0.849)
Dynamic -0.044(0.808) 0.177**(2.316) -0.074(1.163) -0.028(0.356) -0.017(0.118) -0.142*(1.819)
Nature of Data:
Time-Series -0.246**(2.442) -0.226**(2.207) 0.817**(2.509) -0.498***(4.914) -0.186(0.755) 0.522(0.686)
Cross-Sectional 0.062(0.317) 0.314**(2.499) 0.522**(2.386) -0.018(0.096) 0.413***(2.852) 0.219(0.922)
Median Year 0.003(1.126) -0.002(0.730) -0.004*(1.739) 0.006(1.623) -0.002(0.446) -0.003(0.863)
Data Aggregation:
Multiple Countries -1.469***(3.876) -1.011(1.413)
Country 0.515**(2.102) 0.324*(1.657) -0.830**(2.511) 0.708***(3.146) 0.442(1.339) -0.500(0.742)
Region of Country 0.642*(1.952) 0.558***(3.468) 1.405***(4.011) 0.646(1.553) 0.694***(3.433) 0.886*(1.955)
City 0.118(0.618) -0.063(0.722) 0.120(0.551) -0.096(0.568)
Firm -0.708***(2.661) -0.436(1.122)
Estimation Method:
2SLS -0.103(0.382) -0.044(0.278) -0.333*(1.876) -0.383(1.388) -0.011(0.084) -0.266(1.618)
3SLS -0.616***(3.489) -0.161(1.448) -0.120(0.830) -0.393(1.610) -0.161(1.178) -0.227(1.359)
FIML -0.162**(2.365) -0.419***(2.686) 0.106(0.951) -0.103(0.872) -0.532***(2.705) 0.079(0.308)
MLE 0.003(0.063) -0.161(1.229) 0.003(0.052) -0.026(0.212) -0.082(0.471) -0.024(0.232)
Variable OLS Random Effects
North America Asia Europe North America Asia Europe
SUR 0.013(0.233) -0.101(0.889) 0.009(0.150) -0.011(0.123) -0.018(0.136) 0.042(0.553)
GMM 0.398***(3.461) 0.021(0.145) 0.103(0.525) 0.091(0.417)
GLS -0.831***(2.958) 0.479***(4.386) -0.001(0.005) -0.455(1.068) 0.543***(4.630) -0.101(0.536)
Other Method -0.342**(2.329) -0.331(1.156) -0.478**(2.541)
Publication:
Top 36 Journal 0.211***(3.551) 0.168(0.919) -0.202(1.603) 0.287***(2.627) 0.042(0.194) -0.185(0.720)
AJAE 0.121**(2.353) 0.075(0.707) 0.129(1.582) 0.093(0.960) -0.003(0.017) -0.125(0.664)
Book -0.055(0.748) -1.500***(2.788) 0.275***(4.418) -0.052(0.260) -1.327**(2.367) 0.028(0.134)
R2 0.27 0.10 0.24
X2 (1 df) 308.16 47.55 50.72
N 1672 1020 1063 1672 1020 1063