Tangible investment and labour productivity: Evidence from European manufacturing

Figures & data

Table 1. Summary statistics for the variables.

Variable	Mean	Median	Minimum	Maximum	Std. Dev.	IQ range
Apparent labour productivity (thousand euro)	54.5	48.4	5.0	441.7	47.7	53.7
Gross investment in tangible goods per person employed (thousand euro)	8.3	7.2	2.1	149.3	9.0	4.9
Gross investment in land per person employed (thousand euro)	0.2	0.1	0.0	1.0	0.2	0.1
Gross investment in existing buildings and structures per person employed (thousand euro)	0.5	0.2	0.0	10.5	1.2	0.2
Gross investment in construction and alteration of buildings per person employed (thousand euro)	1.3	1.2	−0.0	4.2	0.6	0.7
Gross investment in machinery and equipment per person employed (thousand euro)	5.6	5.1	1.1	18.8	2.9	4.4
Net investment in tangible goods per person employed (thousand euro)	7.2	6.0	−8.2	148.4	10.0	4.9

Table 2. The difference between the real A.L.P. and its forecast.

Country	Real ALP in 2015	Difference	Country	Real ALP in 2015	Difference
Bulgaria	11.40	15.4%	Italy	58.84	−8.1%
Romania	12.76	16.7%	France	71.70	−5.9%
Latvia	17.36	9.0%	Germany	73.55	−7.3%
Lithuania	19.06	10.3%	Finland	74.03	−22.9%
Croatia	19.22	−18.3%	Luxembourg	78.06	−18.0%
Estonia	24.59	17.8%	Austria	82.62	−6.1%
Poland	24.83	−20.5%	U.K.	88.96	17.5%
Slovakia	27.25	3.7%	Denmark	89.32	9.5%
Czech Republic	28.31	2.3%	Sweden	90.07	12.7%
Portugal	28.69	−16.3%	Netherlands	93.10	−2.8%
Hungary	30.56	−8.4%	Norway	96.62	−11.0%
Cyprus	31.40	−28.4%	Belgium	103.51	3.6%
Greece	35.28	−30.9%	Switzerland	150.86	38.5%
Slovenia	37.33	−1.8%	Ireland	441.73	207.5%
Spain	57.62	−12.5%

Table 3. The results of the correlation analysis.

Indicator	G.I.T.G./P.E.	G.I.L./P.E.	G.I.E.B.S./P.E.	G.I.C.B./P.E.	G.I.M.E./P.E.	N.I.T.G./P.E.
Correlation	0.8815	0.2337	0.4675	0.3321	0.8326	0.7961
p-value	0.0000	0.0009	0.0000	0.0000	0.0000	0.0000

Table 4. The results of Granger causality test: p-values.

Indicator (I)	H:*	l = 1	l = 2	l = 3	l = 4	l = 5
G.I.T.G./P.E.	I→ALP	0.4861	0.0000	0.0000	0.0000	0.0000
	ALP→I	0.0000	0.0000	0.0000	0.0000	0.0000
G.I.L./P.E.	I→ALP	0.4631	0.1474	0.7686	0.1466	0.0141
	ALP→I	0.0327	0.4530	0.1097	0.1736	0.1299
G.I.E.B.S./P.E.	I→ALP	0.0005	0.0030	0.0000	0.0000	0.0000
	ALP→I	0.1756	0.0225	0.0015	0.0338	0.0012
G.I.C.B./P.E.	I→ALP	0.6302	0.5431	0.9581	0.0431	0.0702
	ALP→I	0.0222	0.0204	0.0251	0.6063	0.1068
G.I.M.E./P.E.	I→ALP	0.0236	0.0755	0.0389	0.4535	0.0091
	ALP→I	0.1598	0.1916	0.3143	0.2229	0.0606
N.I.T.G./P.E.	I→ALP	0.0000	0.1818	0.1130	0.7654	0.0172
	ALP→I	0.0000	0.2050	0.0277	0.0201	0.2787

Note: * the hypothesis that I does not Granger-cause A.L.P. (I→ALP) is tested in the first row, while the hypothesis that A.L.P. does not Granger-cause I (A.L.P.→I) is tested in the second row.

Table 5. The results of the correlation and causality analyses upon exogenous variables.

Indicator (I)	Correlation analysis between A.L.P. and indicators		The results of Granger causality test: p-values
			H:*	l = 1	l = 2	l = 3	l = 4	l = 5
G.D.P.	correl	0.9369	I→ALP	0.3404	0.0000	0.0000	0.0000	0.0000
	p-value	0.0000	I→GITG/PE	0.0002	0.5933	0.3786	0.0501	0.1403
Export	correl	0.5878	I→ALP	0.4583	0.0781	0.1561	0.3448	0.7108
	p-value	0.0000	I→GITG/PE	0.2511	0.8098	0.8994	0.7837	0.8214
Turnover	correl	0.9549	I→ALP	0.7744	0.0000	0.0000	0.0000	0.0000
	p-value	0.0000	I→GITG/PE	0.0000	0.0000	0.0000	0.0000	0.0000
Cost	correl	0.5780	I→ALP	0.5861	0.3112	0.3965	0.7108	0.9492
	p-value	0.0000	I→GITG/PE	0.7751	0.6252	0.7655	0.7169	0.7713

Note: *the hypothesis that I does not Granger-cause A.L.P. (I→A.L.P.) and G.I.T.G./P.E. (I→G.I.T.G./P.E.) is tested.

Table 6. Probabilities of unit root test (H₀: process has a unit root).

Indicator	Assumes common unit root process (Levin, Lin & Chu)			Assumes individual unit root process (A.D.F.)
	None	With intercept	With trend and intercept	None	With intercept	With trend and intercept
A.L.P.	1.0000	0.1362	0.0000	1.0000	0.9993	0.0005
G.I.T.G./P.E.	0.9139	0.0000	0.0000	0.9935	0.0028	0.0003
G.I.L./P.E.	1.0000	0.0000	0.0000	0.7714	0.0006	0.0001
G.I.E.B.S./P.E.	0.0155	0.0000	0.0000	0.2171	0.0004	0.0036
G.I.C.B./P.E.	0.7329	0.0000	0.0000	0.8918	0.0051	0.0003
G.I.M.E./P.E.	0.3871	0.0168	0.0000	0.9557	0.6294	0.0273
N.I.T.G./P.E.	0.9866	0.0000	0.0000	0.9956	0.0399	0.0017
G.D.P.	1.0000	0.0000	0.0000	1.0000	0.2300	0.0000
Turnover	1.0000	0.0000	0.0000	1.0000	0.0023	0.0179

Table 7. V.A.R. lag order selection criteria.

Lag	L.R.	A.I.C.	B.I.C.	H.Q.
0	–	−0.5854	0.1562	−0.2893
1	191.6991	−3.6214	−2.7562*	−3.2759
2	3.5978	−3.5757	−2.5869	−3.1809
3	16.0773	−3.7511	−2.6387	−3.3069
4	3.6318	−3.7104	−2.4744	−3.2169
5	18.7466*	−3.9575*	−2.5979	−3.4146*
6	5.0508	−3.9498	−2.4667	−3.3576
7	5.5858	−3.9572	−2.3504	−3.3156
8	2.8803	−3.9118	−2.1814	−3.2209

Table 8. The results of the V.A.R. model.

Variable	ln(ALPt)			ln((GITG/PE)t)
	Coefficient	Std. Error	t-statistics*	Coefficient	Std. Error	t-statistics*
C	−0.0732	0.1772	−0.4129	−0.8157	0.6327	−1.2892
ln(ALPt-1)	0.8051	0.1102	7.3056	−1.1887	0.3935	−3.0212
ln(ALPt-2)	0.2382	0.1324	1.7984	3.2291	0.4728	6.8295
ln(ALPt-3)	−0.2076	0.1240	−1.6733	−1.4988	0.4429	−3.3845
ln(ALPt-4)	0.1077	0.1108	0.9721	−0.0746	0.3956	−0.1887
ln(ALPt-5)	0.0158	0.0926	0.1708	−0.6097	0.3306	−1.8443
ln((GITG/PE)t-1)	0.0763	0.0385	1.9845	0.4600	0.1373	3.3499
ln((GITG/PE)t-2)	−0.0112	0.0450	−0.2490	0.1900	0.1608	1.1817
ln((GITG/PE)t-3)	−0.0335	0.0452	−0.7410	0.0468	0.1612	0.2902
ln((GITG/PE)t-4)	0.0133	0.0518	0.2563	−0.4071	0.1849	−2.2015
ln((GITG/PE)t-5)	−0.0605	0.0430	−1.4086	0.4265	0.1534	2.7804
ln(GDPt)	0.3952	0.2077	1.9034	3.2284	0.7414	4.3545
ln(GDPt-1)	−0.4499	0.3023	−1.4885	−3.0825	1.0791	−2.8564
ln(GDPt-2)	0.1184	0.2254	0.5251	−1.3632	0.8048	−1.6938
ln(GDPt-3)	0.0831	0.2275	0.3655	0.9274	0.8122	1.1419
ln(GDPt-4)	−0.2475	0.2121	−1.1670	0.9034	0.7571	1.1933
ln(GDPt-5)	0.1112	0.1317	0.8443	−0.6356	0.4701	−1.3521
ln(Turnovert)	0.6192	0.1146	5.4039	0.0855	0.4091	0.2091
ln(Turnovert-1)	−0.6765	0.1876	−3.6066	−0.0767	0.6697	−0.1146
ln(Turnovert-2)	−0.1201	0.1706	−0.7040	−1.3598	0.6091	−2.2323
ln(Turnovert-3)	0.2437	0.1455	1.6755	1.0918	0.5193	2.1025
ln(Turnovert-4)	−0.0116	0.1290	−0.0901	0.3829	0.4607	0.8310
ln(Turnovert-5)	−0.0202	0.0939	−0.2147	0.2888	0.3353	0.8615
Adjusted R-squared	0.9951			0.8826
Determinant resid covariance	8.17E-05

Table 9. The results of the panel regression analysis considering the impact of G.I.T.G./P.E.

Variable	Coefficient	Std. Error	Prob.
C	−2.4378	1.2554	0.0549
ln(ALPt-1)	0.2806	0.1029	0.0076
ln(ALPt-2)	−0.0515	0.1097	0.6399
ln(ALPt-3)	−0.1524	0.0947	0.1108
ln(ALPt-4)	0.1412	0.0830	0.0922
ln((GITG/PE)t)	0.0118	0.0261	0.6522
ln((GITG/PE)t-1)	0.0189	0.0331	0.5691
ln((GITG/PE)t-2)	0.0566	0.0323	0.0831
ln((GITG/PE)t-3)	0.0034	0.0364	0.9261
ln((GITG/PE)t-4)	−0.0615	0.0360	0.0904
ln(GDPt)	0.2069	0.1694	0.2248
ln(GDPt-1)	−0.0422	0.1787	0.8139
ln(GDPt-2)	−0.1226	0.1736	0.4815
ln(GDPt-3)	0.2753	0.1576	0.0837
ln(GDPt-4)	−0.0200	0.1227	0.8711
ln(Turnovert)	0.7504	0.1066	0.0000
ln(Turnovert-1)	−0.2687	0.1472	0.0709
ln(Turnovert-2)	−0.2019	0.1251	0.1098
ln(Turnovert-3)	0.2300	0.1174	0.0529
ln(Turnovert-4)	−0.0619	0.1092	0.5724
Adjusted R-squared	0.9972
Prob(F-statistic)	0.0000
	Statistic	Prob.
Residual diagnostics: Pesaran C.D.	0.2379	0.8119
Redundant Fixed Effects Tests:
Cross-section F	3.4660	0.0000
Cross-section Chi-square	98.2142	0.0000
Period F	5.4587	0.0000
Period Chi-square	49.4137	0.0000
Cross-section / Period F	4.4599	0.0000
Cross-section / Period Chi-square	138.4492	0.0000

Table 10. The results of the threshold regression.

Variable	Coefficient	Std. Error	Prob.	Coefficient	Std. Error	Prob.
	ALP < 82.54 – 131 obs.			ALP ≥ 82.54 – 23 obs.
C	−0.1124	0.1423	0.4314	−2.2522	2.5583	0.3805
ln(ALPt-1)	0.9296	0.0887	0.0000	0.9723	0.5839	0.0986
ln(ALPt-2)	−0.0211	0.1044	0.8404	−0.8460	0.7984	0.2915
ln(ALPt-3)	−0.1032	0.0968	0.2884	−0.3369	0.3572	0.3475
ln(ALPt-4)	0.1141	0.0788	0.1505	0.6317	0.3457	0.0702
ln((GITG/PE)t)	−0.0018	0.0338	0.9567	0.3420	0.2434	0.1628
ln((GITG/PE)t-1)	−0.0020	0.0375	0.9571	0.0754	0.2081	0.7176
ln((GITG/PE)t-2)	0.0271	0.0366	0.4610	−0.2391	0.1944	0.2212
ln((GITG/PE)t-3)	−0.0179	0.0380	0.6381	−0.1780	0.1599	0.2678
ln((GITG/PE)t-4)	−0.0683	0.0346	0.0511	0.1463	0.1414	0.3031
ln(GDPt)	0.1753	0.1254	0.1648	1.2084	1.1073	0.2775
ln(GDPt-1)	−0.1228	0.1899	0.5190	1.4242	2.3700	0.5491
ln(GDPt-2)	−0.1106	0.1960	0.5738	1.6396	2.3011	0.4776
ln(GDPt-3)	0.1261	0.1812	0.4879	−7.2856	3.2273	0.0259
ln(GDPt-4)	−0.0511	0.1064	0.6317	3.4247	1.5680	0.0310
ln(Turnovert)	0.6151	0.0846	0.0000	0.2285	0.5660	0.6871
ln(Turnovert-1)	−0.8341	0.1276	0.0000	−0.4130	0.8833	0.6410
ln(Turnovert-2)	0.1758	0.1264	0.1671	0.5277	0.5431	0.3333
ln(Turnovert-3)	0.1132	0.1159	0.3309	1.0505	0.3763	0.0061
ln(Turnovert-4)	0.0041	0.0810	0.9595	−1.3894	0.4613	0.0032
Adjusted R-squared			0.9964
Prob(F-statistic)			0.0000
Durbin-Watson stat			1.8249
Heteroskedasticity test: Breusch-Pagan-Godrfrey: Prob. Chi-Square (39)			0.0599

Table 11. The results of the panel regression analysis with consideration of the impact of G.I.E.B.S./P.E. and G.I.M.E./P.E.

Variable	Coefficient	Std. Error	Prob.
C	−1.0624	1.1146	0.3430
ln(ALPt-1)	0.4537	0.1008	0.0000
ln(ALPt-2)	0.0036	0.0925	0.9694
ln((GIEBS/PE)t)	0.0066	0.0081	0.4161
ln((GIEBS/PE)t-1)	0.0072	0.0068	0.2957
ln((GIEBS/PE)t-2)	−0.0080	0.0074	0.2842
ln((GIME/PE)t)	0.0145	0.0221	0.5132
ln((GIME/PE)t-1)	0.0068	0.0323	0.8344
ln((GIME/PE)t-2)	0.0206	0.0292	0.4828
ln(GDPt)	0.2575	0.1914	0.1818
ln(GDPt-1)	−0.0956	0.2053	0.6424
ln(GDPt-2)	0.0885	0.1199	0.4625
ln(Turnovert)	0.3634	0.1208	0.0034
ln(Turnovert-1)	−0.2091	0.1237	0.0944
ln(Turnovert-2)	−0.0542	0.1074	0.6150
Adjusted R-squared	0.9974
Prob(F-statistic)	0.0000
	Statistic	Prob.
Residual diagnostics:
Pesaran C.D.	−1.1822	0.2371
Redundant Fixed Effects Tests:
Cross-section F	2.8891	0.0002
Cross-section Chi-square	73.7865	0.0000
Period F	5.0671	0.0002
Period Chi-square	38.6713	0.0000
Cross-section / Period F	3.6726	0.0000
Cross-section / Period Chi-square	104.5410	0.0000

Table 12. The coefficients of the correlation between A.L.P. and various types of investment per P.E.

Country	G.I.T.G./P.E.	G.I.L./P.E.	G.I.E.B.S./P.E.	G.I.C.B./P.E.	G.I.M.E./P.E.	N.I.T.G./P.E.	A.L.P.
U.K.	0.92***	0.75**	0.54	0.91***	0.96***	0.94***	88.96
Switzerland	0.83**	0.72**	0.74**	−1.00	0.90***	0.71*	150.86
Sweden	0.82***	–	–	0.73*	0.77**	0.82**	90.07
Croatia	0.81***	0.13	0.70**	0.40	0.94***	0.83***	19.22
Hungary	0.81***	−0.15	0.16	0.68**	0.90***	0.81***	30.56
Cyprus	0.80***	0.73**	0.49	0.81**	0.90***	0.88***	31.40
Belgium	0.78***	0.43	0.79**	0.37	0.67*	0.72**	103.51
Netherlands	0.76***	0.66*	0.64*	0.70**	0.59*	0.65*	93.10
Germany	0.71***	0.46	−0.12	0.53	0.64*	0.58	73.55
Estonia	0.70**	0.48	0.37	0.28	0.86***	0.89***	24.59
Luxembourg	0.68**	0.14	0.42	−0.02	0.71**	0.75**	78.06
Poland	0.67**	0.78**	0.13	0.46	0.72**	0.63*	24.83
France	0.64	–	–	–	–	−0.69*	71.70
Ireland	0.61**	−0.41	0.59	−0.04	−0.79**	0.66	441.73
Czech Republic	0.59**	0.18	−0.32	0.08	0.68**	0.59*	28.31
Finland	0.59*	0.83**	0.15	0.11	0.40	0.30	74.03
Greece	0.54	0.26	−0.21	0.75*	−0.06	0.46	35.28
Austria	0.53*	0.02	0.52	−0.44	0.62*	0.41	82.62
Bulgaria	0.47	0.14	0.40	−0.01	0.46	0.39	11.40
Romania	0.35	−0.48	–	−0.05	−0.18	−0.20	12.76
Spain	0.29	0.01	0.67**	−0.49	0.41	0.28	57.62
Portugal	0.20	0.18	−0.63*	0.79**	−0.10	−0.26	28.69
Slovenia	0.07	−0.15	0.54	−0.13	0.59	0.42	37.33
Slovakia	0.07	−0.36	−0.51	−0.60	0.62*	0.70*	27.25
Norway	−0.02	0.94***	–	0.04	0.13	0.01	96.62
Denmark	−0.05	0.12	−0.36	0.11	−0.23	−0.21	89.32
Lithuania	−0.08	−0.23	−0.36	−0.23	0.65*	0.27	19.06
Latvia	−0.11	−0.33	0.08	−0.57	0.11	−0.14	17.36
Italy	−0.19	−0.34	0.17	−0.69*	−0.23	−0.39	58.84

Note: *p < 0.1, ** p < 0.05, *** p < 0.01.