Advertising Economics Jobs / Economist Jobs Help

<!– /* Font Definitions */ @font-face {font-family:”Lucida Console”; panose-1:2 11 6 9 4 5 4 2 2 4; mso-font-charset:0; mso-generic-font-family:modern; mso-font-pitch:fixed; mso-font-signature:-2147482993 6144 0 0 31 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-parent:”"; margin:0in; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:12.0pt; font-family:”Times New Roman”; mso-fareast-font-family:”Times New Roman”; mso-fareast-language:EN-US;} p.MsoFootnoteText, li.MsoFootnoteText, div.MsoFootnoteText {mso-style-noshow:yes; margin:0in; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:”Times New Roman”; mso-fareast-font-family:”Times New Roman”; mso-fareast-language:EN-US;} span.MsoFootnoteReference {mso-style-noshow:yes; vertical-align:super;} @page Section1 {size:8.5in 11.0in; margin:1.0in 1.25in 1.0in 1.25in; mso-header-margin:.5in; mso-footer-margin:.5in; mso-paper-source:0;} div.Section1 {page:Section1;} –>

Section III.A

Results

A summary of the results is presented below in Table III.1, more detailed results are available in Appendix III.A. Observing the table it is clear to see that the values of the coefficients are reasonably different between the two time periods. The fourth column states whether the 1980-1992 period coefficient was more negative, which implies a stronger SRPC relationship. Of the 18 measured, 13 had a stronger (more negative) relationship in the 1980’s than the 1990’s. Whilst this provides evidence for a weakening of the Phillip’s curve relationship in recent years, it is by no means a universal result. Of the five results that are against the trend, three were Scandinavian (Denmark, Finland, and Sweden), the other two were Belgium and Greece.

Country	1990 coefficient	1980 coefficient	More or less negative	R2 1980	R2 1990
Australia	-0.2921132	-0.3144941	More	0.0495	0.229
Austria	N/A	N/A	N/A
Belgium	-0.2803118	0.2192757	Less	0.0288	0.5177
Canada	-0.2758749	-0.6219785	More	0.1692	0.2872
Denmark	-0.0587913	0.4952924	Less	0.0642	0.2266
Finland	-0.1889717	0.5505874	Less	0.0543	0.5296
France	-0.3261998	-2.546941	More	0.8191	0.5048
Germany	-0.7446806	-0.898316	More	0.8421	0.774
Greece	-3.500502	-0.9726864	Less	0.4785	0.797
Ireland	-0.1665611	-1.620649	More	0.7881	0.5699
Italy	-0.4672068	-2.715058	More	0.9079	0.165
Japan	-0.7274278	-3.248461	More	0.8174	0.7437
Netherlands	-0.19729	-0.828843	More	0.5425	0.275
NZ	-0.0801462	N/A	N/A
Norway	-0.0809503	-1.932664	More	0.6805	0.0249
Portugal	-0.955628	N/A	N/A
Spain	0.0041549	-0.8059819	More	0.5719	0.0002
Sweden	-0.65429	-0.5096597	Less	0.0358	0.4727
Switzerland	-0.6544623	-4.325381	More	0.7361	0.4433
UK	0.3375825	-0.745788	More	0.4628	0.1685
USA	0.1854605	-0.0590364	More	0.0032	0.1187
Average	-0.45621	-1.16004		0.447327778	0.380433
Average (without Greece)	-0.29598463	-1.17106448		0.445494118	0.355929

Table III.1

On the whole, approximately two-thirds of the countries experienced a weakening of the unemployment-inflation trade-off. This is further represented by the smaller average coefficient for the 1993-2004 period, falling from -1.16 to -0.46 (or -0.30 without Greece), which is a considerable decrease. Before it took on average a 0.86% movement of unemployment to induce a 1% change of inflation, now it takes over double, specifically a 2.17% (or 3.3% excluding Greece) change of unemployment to induce a 1% change of inflation.

It was of interest to test whether these coefficients were significantly different from each other in order to solidify the change. This was done using a Chow test, the results of which are displayed in Table III.2 (for more detailed data see Appendix III.A)

	Chow Test	Critical value 1%	Critical Value 5%
		5.78	3.47
Australia	37.9937802	Significant	Significant
Belgium	2.60119227	Not Significant	Not Significant
Canada	16.6855827	Significant	Significant
Denmark	10.229843	Significant	Significant
Finland	9.80924029	Significant	Significant
France	27.1391792	Significant	Significant
Germany	5.43887172	Not Significant	Significant
Greece	13.4510827	Significant	Significant
Ireland	53.7109918	Significant	Significant
Italy	9.72249844	Significant	Significant
Japan	12.966096	Significant	Significant
Netherlands	6.58404975	Significant	Significant
Norway	25.6649534	Significant	Significant
Spain	37.0365333	Significant	Significant
Sweden	6.65975593	Significant	Significant
Switz	4.84969637	Not Significant	Significant
UK	15.3281276	Significant	Significant
US	5.35254499	Not Significant	Significant

Table III.2

As the table shows, all but Belgium were significantly different at 5%, and all but Belgium, Germany, Switzerland[1], and US were significantly different at 1%. Consequently it can be stated with greater confidence that there has been a change of the unemployment-inflation trade-off between the two periods, and that the implications need to be explored.

Further, by observing the R² values of the two periods, it can be seen that on average unemployment movements explain inflation movements less in the 1993-2004 period. Although the value is smaller (0.38 in 1993-2004, 0.45 in 1980-2004), it is still clear that inflation changes can be largely explained by unemployment movements; thus it is not that the Phillips’ curve relation has disappeared, instead it has become flatter.

Section III.B

Results

The first result of this section is the aggregate cross section regression of all the countries for the period 1993-2004 with quarterly data. The results are represented by Table III.3 and Figure III.1 below.

Source | SS df MS Number of obs = 885

————-+—————————— F( 1, 883) = 11.63

Model | 13.9439178 1 13.9439178 Prob > F = 0.0007

Residual | 1058.53122 883 1.1987896 R-squared = 0.0130

————-+—————————— Adj R-squared = 0.0119

Total | 1072.47514 884 1.21320717 Root MSE = 1.0949

INFLATION | Coef. Std. Err. T P>|t| [95% Conf. Interval]

————-+—————————————————————-

UNAIRU | -.0712213 .0208828 -3.41 0.001 -.1122071 -.0302356

Table III.3

Figure III.1

As the regression and graph show, the relationship is very weak over the period 1993-2004. With a coefficient of -0.07 the SRPC relationship is practically non-existent, although it is still significant at 1%. Indeed observing the low R² value and how the points are fairly evenly scattered around zero suggests that the unemployment gap has been a poor predictor of inflation changes.

This weak result is surprising given that the unemployment level was shown to still be a decent predictor of inflation in Section III.A. Unfortunately it is difficult to draw a firm conclusion from this result given the drawbacks of the test; these include the fact that the NAIRU values are estimated ex-post and serve little function as a predictor ex-ante; the inaccuracy of the estimates as noted by Staiger, Stock, and Watson (1996); and the fact that the values were interpolated. However, if these pitfalls are ignored it is possible to postulate that the wage and price decisions that make up inflation are near rational, since they are not adjusting to the changes of structural unemployment as the theory suggests, but are based on the unemployment level, or old NAIRU information instead.

Next, this test was performed for each individual country as summarised by Table III.4 (for more detailed regressions see Appendix III.B), which compares the coefficients with the result of Section II.A.

Country	1990 coefficient	1990 U-NAIRU Coefficient	Difference
Australia	-0.2921132	-0.3101632	0.01805
Austria	N/A	-0.0894213
Belgium	-0.2803118	-0.1405221	0.1397897
Canada	-0.2758749	N/A
Denmark	-0.0587913	-0.1155103	0.056719
Finland	-0.1889717	-0.0528902	0.1360815
France	-0.3261998	-0.3014904	0.0247094
Germany	-0.7446806	-0.2717782	0.4729024
Greece	-3.500502	-0.0601352	3.4403668
Ireland	-0.1665611	-0.9254109	0.7588498
Italy	-0.4672068	0.2175667	0.6847735
Japan	-0.7274278	-0.8203434	0.0929156
Netherlands	-0.19729	N/A
NZ	-0.0801462	-0.1410225	0.0608763
Norway	-0.0809503	-0.381808	0.3008577
Portugal	-0.955628	-0.2633415	0.6922865
Spain	0.0041549	0.0362497	0.0320948
Sweden	-0.65429	-0.0448524	0.6094376
Switzerland	-0.6544623	-0.0997716	0.5546907
UK	0.3375825	0.3649101	0.0273276
USA	0.1854605	0.0318187	0.1536418
Average	-0.4562105	-0.177258737	0.4586873
Average (without Greece)	-0.29598463	-0.1837656	0.2832943

Table III.4

As can be the seen the coefficients are generally weakly negative, only Japan and Ireland have values more negative than -0.5, and the average coefficient is -0.18. This average coefficient of -0.18 is small in comparison to the -0.46 of the Section II.A result, but not so different if Greece is considered an outlier and removed. Consequently it is difficult to reaffirm the previous statement made on near rational expectations when this result is in fact quite similar to that of Section II.A, once the individual experiences are observed. Looking at the fourth column it can be seen that the differences of coefficients are small, and on average only 0.28 (without Greece). Indeed the lack of an additional concrete finding here confirms Palley’s (1998) study which demonstrated that NAIRU values tracked unemployment values, and thus the coefficients should not be very different.