: In addressing the issue of short-run heterogeneity as well as long-run homogeneity of the estimated coefficients in a panel framework, the Pooled Mean Group (PMG) estimator (Pesaran et al., 1999) has gained popularity in applied research in economics and business recently. This estimation method has been used successfully in the context of bilateral trade balance estimation for the US with its major trading partners (Goswami and Junayed, 2006), and in another context the bilateral exports and imports model has been extended by modeling bilateral inpayments and outpayments separately for Japan and its major trading partners (Bahmani-Oskooee and Goswami, 2004) by using Autoregressive Distributed Lag (ARDL) approach to cointegration on a partner-by-partner basis. The major limitations of these kind of bilateral models is low power and resulting wrong sign and insignificance of the estimated coefficients due to functional misspecification, multicollinearity as well as omitted variable bias which call for a panel setup in the light of bilateral framework. The present paper fills this gap in the existing literature by estimating both the equations of bilateral inpayments and outpayments for Bangladesh vis-à-vis its 15 major trading partners for the period 1973Q3-2004Q2 using the PMG estimation. The paper reveals that the speed of adjustment measured by the short-run error correction coefficients is lower in PMG estimation compared to ARDL estimation. This might raise another important research question that panel framework may provide better estimates for major parameters of the model in the long run at the cost of allowing lower speed of adjustments in the short run.