M. Hashem Pesaran, Richard J. Smith, and Yongcheol Shin, "Bounds Testing
Approaches to the Analysis of Level Relationships", Journal of Applied
Econometrics, Vol. 16, No. 3, 2001, pp. 289-326.

Bounds Test and ARDL Estimation of the UK Earnings Equation
-----------------------------------------------------------

This file describes the data used in the empirical section of the paper.

There are two ASCII data files, in DOS format, which are zipped in the file
pss-data.zip. Unix users should use "unzip -a".

The file earn1.dat contains the main variables used in the empirical section
of the paper, and the file earn2.data contains the underlying source
variables used in constructing the main variables in earn1.dat.

All data are for the period 1969Q1 to 1997Q4, a total of 116 observations.
In both files, data are arranged by observation. For earn1.dat, there are
two lines for each observation. For earn2.dat, there are nine lines for each
observation. In both files, missing values are denoted by 99999999.9.

The file earn1.dat contains nine variables in the following column order:

	w (real wage), Union, UR (unemployment rate), Prod (labor
	productivity), constant, two dummy variables D7475 and D7579, Wedge
	and time trend.

	In order to replicate estimation, you need to construct first
	differences of main variables, w, Union, UR, Prod and Wedge.

	In particular, the main variables are constructed by

	w = ln(ERPR/PYNONG),
	Wedge = ln(1+TE) + ln(1-TD) - ln(RPIX/PYNONG),
	UR = ln(100ILOU/(ILOU+WFEMP),
	Prod = ln((YPROM+278.29YMF)/(EMF+ENMF)),
	Union = ln(UDEN).
	

The file earn2.dat contains 26 variables in the following column order:

	ERPR, UDEN, GDPMS, FCAS, CGGS, PBRENT, YOG, RXD, GDPA, CGG,
	TYEM, EENIC, WFP, YPROM, YMF, EMF, ENMF, EMPNIC,NIS, OCR,
	ILOU, WFEMP, RPIX, PYNONG, TE, TD.

	Detailed descripitions are given below.

	ERPR:	average private sector earnings per employee (£),
	PYNONG:	the non-oil non-government GDP deflator,
	YPROM:	output in the private, non-oil, non-manufacturing and public
		traded sectors at constant factor cost (£million, 1990),
	YMF:	manufacturing output index adjusted for stock changes
		(1990=100),
	EMF:	employment in UK manufacturing sectors (thousands),  	
	ENMF:	employment in UK non-manufacturing sectors (thousands),
	ILOU:	International Labour Office measure of unemployment
		(thousands),
	WFEMP:	total employment (thousands),
	TE:	the average employers' National Insurance contribution rate,
	TD:	the average direct tax rate on employment incomes,
	RPIX:	the Retail Price Index excluding mortgage payments,
	UDEN:	union density measured as union	membership as a percentage of
		employment (constant from 1980Q4).

	In particular, the following variables are constructed by

	PYNONG = 100(GDPMS-FCAS-CGGS-PBRENT*YOG/(13.102*RXD))/(GDPA-YOG-CGG)
	TE = (EMPNIC+NIS+OCR)/WFP	
	TD = (TYEM+0.914EENIC)/WFP

	where

	GDPMS:	GDP (average measure) at current market prices (£million),
	FCAS:	Factor cost adjustment at current (£million),
	CGGS:	General government consumption (£million),
	PBRENT:	Brent crude oil price ($ per barrel),
	YOG:	Value-added in NS oil and gas extraction (£million,1990),
	RXD:	Sterling dollar cross rate,
	GDPA:	Average measure of GDP at factor cost (£million,1990),
	CGG:	General government consumption (£million, 1990),
	WFP:	Wage and salary bill (including forces' pay) (£million),
	EMPNIC:	Employers' payments of National Insurance Contributions
		(£million),
	NIS:	National insurance surcharge accruals (£million),
	OCR:	Employers' other contributions (£million),
	TYEM:	Acurals of tax on employment income (including PAYE accruing
		on retirement),
	EENIC:	Employees' payments of National Insurance Contributions
		(£million).

Yongcheol Shin
Dept of Economics
University of Edinburgh
50 George Square 
Edinburgh EH8 9YJ, U.K.
Tel: 44 (0)131 650 8351
Fax: 44 (0)131 650 4514
E-mail: Yongcheol.Shin@ed.ac.uk
website: http://www.ed.ac.uk/~shiny/