Estimates of the New Keynesian Phillips Curve for Pakistan
- PDF / 675,500 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 22 Downloads / 195 Views
Estimates of the New Keynesian Phillips Curve for Pakistan Kalim Hyder1 · Stephen G. Hall2,3 Received: 11 December 2017 / Accepted: 26 February 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract This paper presents estimates of the New Keynesian Phillips Curve (NKPC) for the agriculture, manufacturing and services sectors of Pakistan’s economy. The real marginal cost—derived from dynamic translog cost function—labour share of income and output gap are the indicators of economic activity along with past and expected inflation to determine inflation dynamics in each sector. The estimates of the structural parameters of the NKPC are consistent with economic theory in most of the models. Within-sample forecast performance and diagnostic tests indicate that the derived measure of real marginal cost performs better relative to the specifications with labour share of income or output gap. Further, the NKPC based on restrictive Cobb–Douglas production technology with labour input only does not perform better than the models that considers more inputs and intermediate cost. Our results show that the manufacturing is forward-looking sector followed by services and agriculture sectors. Keywords Inflation · Phillips curve · Real marginal cost · Pakistan JEL Classification E31 · E32 · E47 · E52
1 Introduction Empirical investigation of the New Keynesian Phillips Curve (NKPC) has garnered research attention of many researchers (Gali et al. 2005; Christiano et al. 2005; Neiss and Nelson 2005; Batini et al. 2005) especially after the estimation of the structural model of inflation by Gali and Gertler (1999) and Sbordone (2002). The important aspect of Gali and Gertler (1999) is to establish the empirical link between short-run
* Kalim Hyder [email protected] 1
Monetary Policy Department, State Bank of Pakistan, I.I. Chundrigar Road, Karachi, Pakistan
2
The University of Leicester School of Business, University Road, Leicester LE1 7RH, UK
3
University of Pretoria, Lynnwood Rd, Hatfield, Pretoria 0002, South Africa
13
Vol.:(0123456789)
K. Hyder, S. G. Hall
inflation dynamics and real economic activity with strong theoretical underpinnings. They replace the ad hoc measure of economic activity (i.e. output gap) by the labour share of income considering it as a proxy of real marginal cost. Mazumder (2010), however, criticises usage of the labour share of income due to its countercyclical behaviour. Consequently, Dupuis (2004) and Batini et al. (2005) consider a variety of variables that represent the better dimension of real marginal cost. In continuation of this literature, we derive a new measure of real marginal cost from the dynamic translog cost function. Regarding the estimation of NKPC on countries’ aggregate data, Byrne et al. (2013) and Norkute (2015) highlight an important point that if price-setting behaviour differs across sectors then there is a possibility of getting biased parameter estimates in NKPC models that are applied to aggregated data. In such situation, the impor
Data Loading...
