Download 4. S D upply and

Central Bank of the Republic of Turkey
4. Supply and Demand Developments
The first-quarter national accounts data are consistent with the outlook
presented in the April Inflation Report. Economic activity remained robust, albeit
having a slower rate of growth than in the first quarter, while the main driver of
growth was domestic demand, primarily of the private sector. Meanwhile,
exports remained weak and imports continued to accelerate, causing net
external demand to make a negative contribution to growth. Thus, the
divergence between domestic and external demand growth displayed during
the exit phase has become more pronounced, and the foreign trade deficit
widened further.
Second-quarter data indicate a quarter-on-quarter weakening in
economic activity. Seasonally adjusted data showed that industrial production
has contracted for four consecutive months since February, and domestic
demand indicators also confirmed the slowdown. A featured question while
heading into the second half of the year is whether this trend will be permanent
or not. Leading indicators signal that the slowdown in the second quarter will
not turn into a long-term recession. However, given the lagged effects of the
policy measures and the pace of slowdown at the global scale, domestic
demand is expected to settle into a milder path of growth.
Under the current outlook with receding unit labor costs owing to
productivity gains and low levels of capacity utilization rates due to weak
external demand, aggregate demand conditions are not expected to exert an
upward pressure on inflation. However, both the high levels of energy and other
commodity prices as well as the developments that weaken external demand
will delay the recovery of the current account balance.
Inflation Report 2011-III
47
Central Bank of the Republic of Turkey
4.1. Gross Domestic Product Developments and Domestic
Demand
According to the national accounts data released by TurkStat, GDP
increased by 11.0 percent year-on-year during the first quarter of 2011. The
largest contributor
to
annual
growth
was
private
demand,
for
both
consumption and investment. Due to relatively weak exports and strong import
demand, net external demand made a negative contribution to annual growth
(Chart 4.1.1).
The seasonally adjusted GDP expanded by 1.4 percent in the first quarter,
slowing down slightly. Having maintained its strong pace during this period, final
domestic demand was the main driver of the quarterly growth (Chart 4.1.2).
Meanwhile, external demand remained relatively weak, and hence, external
and domestic demand diverged further.
Chart 4.1.1.
Chart 4.1.2.
Contribution to GDP Growth by Demand
Components
GDP and the Final Domestic Demand
(Seasonally Adjusted, 2008 Q1=100)
(Percent)
15
GDP
11.0
10
8.7
Final Domestic Demand
115
7.2
5
0.2
110
1.8
105
0
0.7
-0.3
100
-5
95
GDP
Inventories
Exports
Public Investment
Public Consumption
Private Investment
Private Consumption
Source: TurkStat.
Imports
-7.4
-10
90
85
80
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1
2005
2006
2007
2008
2009
2010 2011
Source: TurkStat, CBRT.
The second-quarter data point to a slowdown of the robust growth in
private sector demand. Production and imports of consumption goods, which
are among private consumption demand indicators, remained below the firstquarter averages in the April-May period (Chart 4.1.3). Automobile and white
goods sales also displayed a quarter-on-quarter decline in the second quarter
(Chart 4.1.4). Consumer confidence indicators indicated a moderate pace for
consumption demand as well (Chart 4.1.5).
48
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Chart 4.1.3.
Chart 4.1.4.
Production and Import Quantity Indices of
Consumption Goods (Seasonally Adjusted, 2005=100)
Domestic Sales of Automobiles and White Goods
Production
(Thousand, Seasonally Adjusted)
120
230
115
110
105
55
190
50
170
45
150
40
130
35
90
100
70
50
95
60
210
110
30
2006
2007
2008
2009
600
550
500
450
30
400
25
350
20
15
300
12341234123412341234123412
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2005
Automobiles
White Goods (right axis)
Imports (right axis)
2005
2010 2011
2006
2007
* As of May.
Source: TurkStat, CNBC-e, CBRT.
Source: AMA,WGIA, CBRT.
Chart 4.1.5.
Chart 4.1.6.
Consumer Confidence
Weekly Consumer Loans
2008
2009
2010 2011
(Weekly Nominal Percent Change, 13-Week Average)
CNBC-e
CBRT (right axis)
Total
Other
1.5
120
100
110
95
100
90
90
85
80
80
70
75
-0.5
60
70
-1.0
Housing
Automobile
1.0
0.5
Source: TurkStat, CNBC-e, CBRT.
0108
0308
0508
0708
0908
1108
0109
0309
0509
0709
0909
1109
0110
0310
0510
0710
0910
1110
0111
0311
0511
0711
0711
0511
0311
0111
1110
0910
0710
0510
0310
0110
1109
0909
0709
0509
0309
0.0
Source: CBRT.
Investment demand indicators also signal a slowdown in economic
activity in the second quarter. The downward trend in capacity utilization rates
in this period partially reduced the need for investments. Imports of investment
goods continued to rise while the production of investments goods declined in
the April-May period compared to the first quarter (Chart 4.1.7). Relative price
movements are considered to account for the post-crisis divergence between
imports and the production of investment goods (Box 4.1). Domestic sales of
light and heavy commercial vehicles remained below the first-quarter averages
(Chart 4.1.8).
Inflation Report 2011-III
49
Central Bank of the Republic of Turkey
Chart 4.1.7.
Chart 4.1.8.
Production and Import Quantity Indices of
Capital Goods (Seasonally Adjusted, 2005=100)
Domestic Sales of Commercial Vehicles
(Thousand, Seasonally Adjusted)
Production
Imports
Production (excl. motor vehicles)
Imports (excl. transport)
220
Light Commercial
Heavy Commercial (right axis)
30
5.0
4.5
200
25
4.0
180
3.5
20
160
3.0
140
15
120
100
2.5
2.0
10
1.5
80
5
60
2006
2007
2008
2009
2010
1.0
1234123412341234123412341234123412
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2003 2004 2005 2006 2007 2008 2009 20102011
2011
* As of May.
Source: TurkStat, CBRT.
Source: AMA, CBRT.
A featured question at this point is whether the second quarter slowdown
in domestic demand indicators will become permanent or not. The slowdown in
the second quarter is attributed to the recently adopted policy measures as
well as the general election process and the deteriorating global growth
outlook. The ongoing increase in consumer loans during the first two weeks of
June, albeit a slight slowdown, indicates that the credit channel continues to
support growth (Chart 4.1.6). Having stabilized at high levels since the onset of
2011, 12-month-ahead investment expectations maintained this trend also in
June, signaling that the investment propensity remains strong (Chart 4.1.9). As of
May, the composite index constructed by aggregating selected leading
economic indicators hardly contains signals for a permanent slowdown
(Chart 4.1.10).
Chart 4.1.9.
Chart 4.1.10.
12-Month Ahead BTS Expectations for Investment
Leading Indicators Index
(Up-Down, Seasonally Adjusted)
(Seasonally Adjusted)
40
104
30
102
20
10
100
0
98
-10
96
-20
-30
94
-40
92
-50
50
0611
0311
1210
0610
0910
1209
0310
0909
0309
0609
0608
0908
1208
1207
0308
0307
0607
0907
Source: CBRT.
0597
0198
0998
0599
0100
0900
0501
0102
0902
0503
0104
0904
0505
0106
0906
0507
0108
0908
0509
0110
0910
0511
90
-60
Source: TurkStat, CBRT.
Inflation Report 2011-III
Central Bank of the Republic of Turkey
In sum, given the recently released data, it is estimated that the domestic
demand declined in the second quarter (Chart 4.1.11). However, both the
general election process and the escalating global problems in the said period
complicates to monitor the underlying trend in domestic demand and to
identify the effects of the policy measures. Current indicators signal that the
contraction in the second quarter is not permanent. However, given the lagged
effects of the policy measures and the global slowdown, domestic demand is
expected to follow a milder path of growth in the second quarter.
Chart 4.1.11.
Final Domestic Demand
(Seasonally Adjusted, 2008Q1=100)
115
110
105
100
95
90
85
80
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2005
2006
2007
2008
2009
2010
2011
* Estimate.
Source: TurkStat, CBRT.
4.2. External Demand
The first-quarter outlook for external demand was broadly consistent with
the April Inflation Report forecasts. While exports of goods and services
increased by 7.7 percent year-on-year, imports of goods and services were up
27.0 percent in this quarter, resulting in a further negative contribution of net
external demand to annual growth (Chart 4.2.1). In seasonally adjusted terms,
exports followed almost a horizontal course, and remained below pre-crisis
levels. Meanwhile, imports displayed a robust quarter-on-quarter increase.
Reflecting the divergence between demand components, this situation has led
to a deterioration in foreign trade balance (Chart 4.2.2.)
Inflation Report 2011-III
51
Central Bank of the Republic of Turkey
Chart 4.2.1.
Chart 4.2.2.
Contribution of Net External Demand to Annual
GDP Growth
Exports and Imports of Goods and Services
(Seasonally Adjusted, 1998 Prices, Billion TL)
(Percent)
6
Exports
Imports
Net Exports
Exports
Imports
9
4
8.5
8
2
7.5
0
7
-2
6.5
-4
6
-6
5.5
-8
5
1
2009 2010
2
3
2010
4
1
2*
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2011
* Estimate.
Source: TurkStat, CBRT.
2005
2006
2007
2008
2009
2010 2011
* Estimate.
Source: TurkStat, CBRT.
The seasonally adjusted quantity index excluding gold, one of the key
indicators of exports, maintained its gradual increase in the second quarter of
2011 (Chart 4.2.3). Compared to the pre-crisis period, exports still display a
weaker pace of recovery and continue to dampen aggregate demand. Given
the recently released data, exports of goods and services are estimated to
have increased at a modest pace in the second quarter (Chart 4.2.2).
Chart 4.2.3.
Chart 4.2.4.
Quantity Index for Exports Excluding Gold
Imports and Industrial Production Indices for the
Global Economy
(Seasonally Adjusted, 2003=100)
(Seasonally Adjusted, 2005=100)
Imports
190
Industrial Production
125
120
170
115
150
110
105
130
100
110
95
90
90
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 12*
2003 2004 2005 2006 2007 2008 2009 20102011
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2005
2006
2007
2008
2009
2010 2011
* As of May.
* As of May.
Source: TurkStat, CBRT.
Source: Netherlands Bureau for Economic Policy Analysis.
Recent developments in international markets have increased downside
risks to global growth. The possible spillover of the Greek debt crisis into other
52
Inflation Report 2011-III
Central Bank of the Republic of Turkey
European countries, the slowing pace of recovery in the U.S. and the weak
Japanese economy following the earthquake led the economic activity to slow
down on a global scale (Chart 4.2.4). Accordingly, growth forecasts were
revised downwards especially for the U.S. and the indebted Greek, Portuguese
and Irish economies, compared to the previous reporting period (Chart 4.2.5).
The global manufacturing and services PMI indices suggest that the slowdown
in economic activity may continue into the third quarter of the year
(Chart 4.2.6). Meanwhile, growth forecasts for the euro area, our main trading
partner, remained virtually unchanged, barring a remarkable revision to the
external demand outlook. As a result, the prediction that the recovery in exports
will be slow and gradual parallel to the economic developments in the external
markets is maintained in this reporting period as well.
Chart 4.2.5.
Chart 4.2.6.
GDP-Weighted Global Production Index
Global PMI Indices
(Seasonally Adjusted, 2009Q1=100)
(Seasonally Adjusted)
April 2011
Manufacturing
July 2011
115
Services
65
60
110
55
50
105
45
40
100
35
95
Source: Bloomberg, CBRT.
1
2
3
2012
4
0611
2011
4
1110
3
0410
2
0909
1
0209
2010
4
0708
3
1207
2009
2
0507
1
1006
4
0306
3
0805
2
0105
30
1
Source: Bloomberg.
Due to the robust course of domestic demand in the first quarter of
2011, imports increased sharply. Second-quarter indicators point to a weaker
course for domestic demand and imports compared to the previous quarter. In
fact, the seasonally adjusted import quantity index remained below the firstquarter average in the April-May period (Chart 4.2.7). In this scope, imports of
goods and services are estimated to have decreased in the second quarter on
a quarterly basis (Chart 4.2.2). An analysis of main industrial groups suggests that
imports of consumption and investment goods are well above pre-crisis levels
parallel to the domestic demand, while imports of intermediate goods remain
limited in line with the weak course of external demand. This observation is
another indicator of the divergence between domestic and external demand
components (Chart 4.2.8).
Inflation Report 2011-III
53
Central Bank of the Republic of Turkey
Chart 4.2.7.
Chart 4.2.8.
Quantity Index for Imports
Quantity Indices for Imports by Subcategories
(Seasonally Adjusted, 2003=100)
(Seasonally Adjusted, 2008Q1=100)
Investment Goods
Intermediate Goods
Consumption Goods
200
150
190
135
180
120
170
105
160
90
150
75
140
130
60
120
45
110
30
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2005
2006
2007
2008
2009
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2010 2011
* As of May.
Source: TurkStat, CBRT.
2005
2006
2007
2008
2009
2010 2011
* As of May.
Source: TurkStat, CBRT.
In sum, parallel to the mild recovery in exports and the slowdown in the
demand for imported goods, the negative contribution of net external demand
to growth is expected to have declined in the second quarter of the year
(Chart 4.2.1). However, high energy and other commodity prices coupled with
developments weakening external demand delay the recovery of the current
account balance. In fact, the recently released data suggest that the
deterioration in the foreign trade balance still persists in the second quarter of
2011, albeit at a slower pace (Chart 4.2.9). Given the current global outlook of
weak external demand conditions, containing domestic demand is still critical
with respect to both inflation outlook and financial stability.
Chart 4.2.9.
Current Account Balance
(Seasonally Adjusted, Billion USD)
Current Account (excl. energy)
10
Current Account
5
0
-5
-10
-15
-20
-25
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
2003
2004
2005
2006
2007
2008
2009
2010 2011
* Estimate for June.
Source: TurkStat, CBRT.
54
Inflation Report 2011-III
Central Bank of the Republic of Turkey
4.3. Labor Market
First-quarter employment developments turned out to be more favorable
than the outlook presented in the April Inflation Report. The ongoing uptrend in
non-farm employment since the last quarter of 2010 was maintained in the
January-April period. Farm employment, which displayed a similar course to
non-farm employment up to April, slowed down in this period (Chart 4.3.1). With
the sharp increase in employment, the unemployment rate has fallen back to its
pre-crisis levels as of the first quarter of 2011 (Chart 4.3.2).
Chart 4.3.1.
Chart 4.3.2.
Farm and Non-Farm Employment
Unemployment
(Seasonally Adjusted, Million)
(Seasonally Adjusted, Percent)
Labor Force Participation Rate (right axis)
Unemployment Rate
Non-Farm Unemployment Rate
Non-Farm Employment
Farm Employment (right axis)
18.0
8.0
17.5
7.5
17.0
7.0
16.5
6.5
16.0
20
51
5.5
12
47
15.0
5.0
10
46
14.5
4.5
14.0
4.0
8
45
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2*
0411
1010
15.5
0410
48
1009
14
0409
6.0
1008
49
0408
16
1007
50
0407
18
2007
2008
2009
2010
2011
* As of April.
Source: TurkStat, CBRT.
Source: TurkStat, CBRT.
The increase in non-farm employment in the first quarter of 2011 was
provided through the contribution by all main sectors, primarily the services and
the industrial sector (Charts 4.3.3 and 4.3.4). However, in April, the services
sector continued to support non-farm employment growth, while the robust rate
of increase in the industrial sector slowed down.
Chart 4.3.3.
Chart 4.3.4.
Industrial Employment and Production
Services and Construction Sector Employment
(Seasonally Adjusted)
(Seasonally Adjusted, Million)
Industrial Employment
Industrial Production (right axis)
Million
Construction
Source: TurkStat, CBRT.
Inflation Report 2011-III
10.2
95
1.2
10.0
90
1.0
9.8
Thousands
1.4
0411
10.4
100
1210
1.6
0810
10.6
105
0410
1.8
1209
10.8
110
0809
2.0
0409
0511
0111
0910
0510
0110
0909
0509
0109
0908
0508
0108
0907
3.8
11.0
115
1208
4.0
2.2
0808
4.2
11.2
120
0408
4.4
11.4
2.4
1207
4.6
2.6
125
0807
4.8
130
0407
5.0
0507
Services (right axis)
2005=100
Source: TurkStat, CBRT.
55
Central Bank of the Republic of Turkey
Recent developments suggest that the rate of increase in industrial
employment may decelerate on the back of the slowdown in economic
activity. Industrial production contracted for four consecutive months in the
February-May period, while the PMI employment index, a leading indicator for
employment developments, maintains its low level compared to the first
quarter, albeit having increased slightly in June. (Chart 4.3.5). These indicators
point that employment conditions in the industrial sector may deteriorate as of
May.
The recovery in employment conditions in the first quarter of 2011
continued to support domestic demand (Chart 4.3.6). Despite going down to
pre-crisis levels, unemployment rates are unlikely to have exerted remarkable
pressure on wages in the said period. In fact, no significant upward movement
was observed on labor costs in the first quarter of 2011, and the real wage index
reflecting developments in average hourly wages remained flat (Chart 4.3.7).
On an annual basis, productivity gains exceeded the increase in real wage
index, and thus, real unit wages fell back to pre-crisis levels (Chart 6, Box 4.2).
However, the fact that the current data on average wage developments
reflecting labor costs are also influenced by the employment composition and
the lack of a wage index for showing only wage movements, obstructs a sound
analysis on wages (Box 4.2).
Chart 4.3.5.
Chart 4.3.6.
Manufacturing Industry Employment
(Quarterly Percent Change) and PMI Employment
Index
Household Spending and Real Wage Payments*
(Seasonally Adjusted, 2007=100)
(Seasonally Adjusted)
Real Wage Payments-Short-Term Labor Statistics
Manufacturing Industry Employment Index (ILII)
Consumption Spending (excl. furniture, household appliances and
maintenance, transport and communication)
PMI (right axis)
6
65
110
4
60
105
2
55
0
50
-2
45
-4
40
-6
35
85
-8
30
80
100
0611
1210
0610
1209
0609
1208
0608
1207
0607
95
90
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1
2005
2006
2007
2008
2009
2010 2011
* Calculated by the weighted average of total wages paid in industrial,
construction, trade, accommodation-catering services, transportwarehousing sectors. Deflated by CPI.
Source: TurkStat, Markit, CBRT.
56
Source: TurkStat, CBRT.
Inflation Report 2011-III
Central Bank of the Republic of Turkey
To sum up, non-farm employment maintained its uptrend, while industrial
employment displayed a slight decline as of April 2011. Leading indicators
suggest that employment conditions in the industrial sector may deteriorate
starting from May (Chart 4.3.8). In addition, both the stabilizing policy measures
as well the envisioned milder course of aggregate demand conditions in the
second quarter on the back of the recently escalating global problems are
expected to slow down the rate of employment growth compared to the past
two quarters.
Chart 4.3.7.
Chart 4.3.8.
Hourly Labor Cost
Non-Farm Value Added and Employment
(Seasonally Adjusted, 2008=100)
(Seasonally Adjusted)
Labor Earnings (annual percent change, right axis)
Labor Earnings
Real Labor Earnings*
130
125
120
115
110
105
100
95
90
85
80
2007
2008
Inflation Report 2011-III
2009
2010
2011
1998 Prices
Billion TL
Million
26
18.0
12
25
17.5
24
17.0
23
8
22
6
21
4
* Deflated by CPI.
Source: TurkStat, CBRT.
Employment (right axis)
14
10
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1
Value Added
16.5
16.0
15.5
20
19
15.0
2
18
14.5
0
17
14.0
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 12*
2005
2006
2007
2008
2009
2010 2011
* Estimate.
Source: TurkStat, CBRT.
57
Central Bank of the Republic of Turkey
Box
Prices of Investment Goods and Investment Spending
4.1
Displaying
a rapid recovery in the post-crisis period, private sector machinery
and equipment investments gained momentum in the last quarter of 2010, and
also continued to rise in the first quarter of 2011. Thus, private sector machinery
and equipment investments played a great role in the robust recovery of the
domestic demand during the post-crisis period (Chart 1). The strong course of
private sector machinery and equipment investments is attributed to favorable
demand expectations and better financing conditions as well as the relatively
lower prices for investment goods. In fact, an analysis of the relative prices
obtained from dividing the investment goods deflator by the GDP deflator
suggests that the relative price of investment goods trended downwards in the
post-2003 period (Charts 2 and 3). Meanwhile, the decline in the relative prices of
investment goods causes the share of private sector machinery and equipment
investments within GDP to vary at constant and current prices (Chart 4). This
divergence became more pronounced after end-2003, when relative prices
started to decline. Prices of investment goods that increased more moderately
than general prices indicate that more physical value can be obtained by
allocating a lower share of the GDP. In fact, in terms of the share within GDP,
machinery and equipment investments, which approached pre-crisis levels at
current prices in the first quarter of 2011, realized well above pre-crisis levels at
constant prices.
Chart 1. Contribution of Private Sector
Machinery/Equipment Investments to Annual
Growth and GDP Growth (Percent)
(Seasonally Adjusted)
GDP Deflator
Contribution of Private Mach.Eq.Inv.
GDP Growth
14
Chart 2. Private Sector Machinery/Equipment
Investments and GDP Deflators
12
Private Mach.Eq. Inv. Deflator
12
11
10
10
8
6
9
4
2
8
0
7
-2
6
-4
5
Source: TurkStat.
58
2011-1
2010-4
2010-3
2010-2
2010-1
2010
2009
2008
2007
2006
2005
2004
-6
123412341234123412341234123412341
2003 2004 2005 2006 2007 2008 2009 20102011
Source: TurkStat, CBRT.
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Chart 3. Relative Prices of Investment Goods
and the Share of Private Sector
Machinery/Equipment Investments within GDP
Chart 4. The Share of Private Sector
Machinery/Equipment Investments within GDP
(Seasonally Adjusted)
(Seasonally Adjusted)
Private Mach.Eq. Inv./GDP (constant prices)
Private Mach.Eq. Inv. Deflator/GDP Deflator
Private Mach.Eq. Inv./GDP (constant prices,
right axis)
19
1.2
19
Private Mach.Eq. Inv./GDP (current prices)
17
17
Source: TurkStat, CBRT.
The
2011-1
2010-2
2009-3
2008-4
2008-1
2007-2
2006-3
2005-4
2005-1
2004-2
2003-3
2002-4
2011-1
2009-3
2010-2
2008-4
2007-2
2008-1
2005-4
2006-3
2004-2
2005-1
2002-4
2003-3
5
2001-2
2002-1
7
5
2000-3
7
0.6
1999-1
1999-4
9
1998-2
9
0.7
2002-1
11
2001-2
11
2000-3
0.8
13
1999-4
13
0.9
15
1999-1
15
1.0
1998-2
1.1
Source: TurkStat, CBRT.
fall in the relative prices of imported investment goods is also believed to
have an impact on the relative cheapening of investment goods. In fact, prices
of imported investment goods diverged from overall import prices to a great
extent after 2003 (Chart 5). In this respect, it would be helpful to consider that,
among other factors, the price advantage has also contributed to the increase in
imports of investment goods (Chart 6).
Chart 5. Imports Unit Value Indices
(2003=100)
Chart 6. Investment Goods Imports
(Billion USD)
Unit Value of Capital Goods Imports
205
Unit Value of Overall Imports
Parts of Capital Goods
Capital Goods
50
45
185
40
165
35
30
145
25
20
125
15
105
10
5
85
123412341234123412341234123412341
2003 2004 2005 2006 2007 2008 2009 20102011
Source: TurkStat.
0
2004 2005 2006 2007 2008 2009 2010 2011*
* Estimate by compiling 2010 June- 2011 May data.
Source: TurkStat.
In sum, the decline in relative prices of investment goods in recent years caused
the relative cost of investments to fall, and supported the growth of private sector
machinery and equipment investments amid rising investment appetite and
favorable financing conditions. The surge in investments fed into macro financial
risks in the short term as it deteriorates the current account balance; however, it
will positively contribute to both inflation and potential production by expanding
the productive capacity of the economy in the medium term.
Inflation Report 2011-III
59
Central Bank of the Republic of Turkey
Box
Data on Wages and Earnings
4.2
Developments on wages and productivity as well as the labor market structure
are significant factors affecting inflation through both demand and cost
channels. In periods when aggregate demand remarkably exceeds the potential
output (periods of overheating in the economy), wages may be subject to
pressure and the medium-term inflation outlook may deteriorate. Thus, the course
of wages is critical to central banks and financial markets. Several wage series
based on various methods and criteria are present for Turkey. In order to be able
to interpret these series from an economic perspective, understanding how these
series are derived is crucial. This Box analyzes various data on wages and earnings
published by different sources. Moreover, these indicators are compared on a
sectoral basis to reveal their similarities and differences.1
Sources
regarding wage series can be classified under four main categories:
hourly costs and earnings indices published under Labor Cost Indices, wages per
hour worked indices calculated indirectly from Short-Term Business Statistics, Labor
Cost Index published under Building Construction Cost Indices, and average daily
earnings obtained from SSI bulletins. The first three of these series are announced
by TurkStat on a quarterly basis, while the last one is a monthly publication by the
SSI.2
The Labor Cost Index, which has been
published since July 2010, measures
the nominal hourly cost of employing
Chart 1. Non-Farm Costs and Earnings
(Real, 2008=100, Seasonally Adjusted)
110
Cost
108
Earnings
106
a wage earner. The main components
104
102
of this index are divided into two as
earnings
(the
changes
in
indicator
regular
or
of
hourly
irregular
payments made to wage earners)
and
non-earnings
(social
security,
severance and termination payments
100
98
96
94
92
90
I
II III IV I
II III IV I
II III IV I
II III IV I
2007
2008
2009
2010
2011
Source: TurkStat Labor Cost Index, CBRT.
by the employer). The difference between the two series, which move in parallel,
1
All data, excluding those from the SSI, are seasonally adjusted. Data are deflated by CPI.
the rest of the Box, hourly wages calculated from Short-Term Business Statistics will be referred to as wages per hour worked
to avoid confusion with those from the Labor Cost Index.
2 In
60
Inflation Report 2011-III
Central Bank of the Republic of Turkey
is affected by legal arrangements that are reflected on costs. In fact, the
difference between costs and earnings indices grew stronger due to legal
arrangements on premium payments of employers introduced at end-2008
(Chart 1).3
Another series to monitor sectoral average wage developments is the Short-Term
Business Statistics published by the TurkStat. This dataset has been published since
2005 on a quarterly basis for the industrial, construction and services sectors
compiling employment, hours worked and gross wages-salaries indices.
Short-Term
Business Statistics and Labor Cost Indices have been published since
2005 and 2007, respectively. Although the data sources of Short-Term Business
Statistics and Labor Cost Indices are mostly similar, differences exists with respect
to calculation of these indicators. Main differences can be listed as follows:4 Firstly,
there are no hourly wage series directly published under Short-Term Business
Statistics. For each main sector, the gross salaries index that reflects total wage
payments is divided by hours worked in order to calculate the wage per hour
worked. In the Labor Cost Index on the other hand, hourly wage data are directly
published by main sectors and sub-sectors (sections).5
Another difference is in weighting.
While constructing indices for main sectors,
sub-indices are aggregated by using the sectoral weights obtained from the base
year values of the respective variable. The base years used for sectoral weight
calculations of these indices differ. The base years for the indices published under
Short-Term Business Statistics and Labor Cost Indices are 2005 and 2008,
respectively. Moreover, gross wages-salaries and hours worked indices, which are
components of the wage per hours worked index, are aggregated by their own
weights, while the hourly labor cost calculated at sectional level is aggregated by
weights of gross wages. 6
3
As per the Law on Amendment to the Labor Act No 763 and Some Acts, private sector employers who employ workers were
provided with incentives. Accordingly, 5 points of the employers’ share in insurance premiums shall be covered by the Treasury.
This amendment was put into effect on October 1, 2008.
4 These differences are compiled by the announcements of the TurkStat.
5 The Statistical Classification of Economic Activities in the European Community (NACE), which serves as a basis for TurkStat’s
data, runs from general to specific as Section, Division, Group and Class. For example, the NACE Rev2 classification is
composed of 21 Sections, 88 Divisions, 272 Groups and 615 Classes.
6 Hourly wage indices are calculated under the labor cost indices on a sectional basis. Then, aggregation is made by using
base year weights obtained from gross wage-salary values. Under Short-Term Business Statistics, employment, hours worked
and gross salaries-wages indices calculated at class or group levels are aggregated by main sectors using base year weights
of these variables.
Inflation Report 2011-III
61
Central Bank of the Republic of Turkey
In addition, differences between these indices also exist in terms of scope. In the
calculation of Labor Cost Indices, data obtained from enterprises with twenty or
more employees in industrial and construction sectors are adjusted to cover
enterprises with one or more employees by using the coefficients from Annual
Labor Statistics. Gross salary-wage and hours worked indices published under the
Short-Term Business Statistics are not adjusted as above.
Another
difference regarding the coverage is the exclusion of finance and
insurance activities from the services sector compiled under Short-Term Business
Statistics. The said sub-sector is included in the Labor Cost Index by using data
obtained from the BRSA.
To sum up, hourly cost and earnings indices are considered to be more reliable
than wage per hours worked indices as they are computed at a more
disaggregated level.
Hourly
cost and earnings indices and wages per hour worked indices are
consistent in general across sectors (Charts 2-4). As the gross wage-salary index
under the Short-Term Business Statistics does not include the insurance premium of
the employee paid by the employer, the level of the wage per hour index is
closer to that of the hourly earnings index. The difference between these indices is
more evident for the services sector (Chart 2). It is believed that differences arising
from the above-mentioned aggregation are most prominent in the services sector
due to higher number of sub-items in this sector.
Another
data for monitoring average wage developments are found in the
Insured Person Statistics Bulletin published by the SSI. These data are calculated by
the reported daily earnings applicable to employer-sponsored insurance
premiums. Although, the availability of these data on monthly frequency and
sectoral basis is an advantage, the irregular publication of SSI bulletins and the
high volatility of daily average earnings limit the use of the data. Due to this
volatility, converting the data to a quarterly series by taking 3-month averages
enhances the informative value. The graphical representation of this series with
labor earnings and average wage per hours worked indices displays a similar
pattern, notwithstanding the absence of a one-to-one relation (Charts 2-4).
Moreover, SSI earnings data are also parallel with minimum wage, which sets a
lower bound to wages (Chart 5).
62
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Chart 2. Services Wage Indices
Chart 3. Construction Wage Indices
(Real, 2008=100, Seasonally-Adjusted)
(Real, 2008=100, Seasonally-Adjusted)
Cost
Wage/Hour
Earnings
SSI Earnings
115
Cost
Wage/Hour
Building Construction Cost Index
Earnings
SSI Earnings
120
115
110
110
105
105
100
100
95
95
90
I II III IV I II III IV I II III IV I II III IV I
2007
2008
2009
2010
I II III IV I II III IV I II III IV I II III IV I II
2011
2007
2008
2009
2010
2011
Source: TurkStat, SSI, CBRT.
Source: TurkStat, SSI, CBRT.
Chart 4. Industrial Wage Indices
Chart 5. Average Daily Earnings and Minimum
Wages in Non-Farm Sectors
(Real, 2008=100, Seasonally Adjusted)
(Real, 2008=100)
Cost
Wage/Hour
Earnings
SSI Earnings
110
Minimum Wage
110
SSI Earnings
108
105
106
104
100
102
100
95
98
90
The
0211
1010
0610
0210
1009
0609
0209
1008
2011
0608
2010
0208
2009
1007
2008
0607
2007
Source: TurkStat, SSI, CBRT.
0207
96
I II III IV I II III IV I II III IV I II III IV I
Source: SSI, CBRT, Ministry of Labor and Social Security.
wage indices mentioned so far reflect average wage developments. An
important point to underline at this point is that the change in indices is driven by
both wage movements and changes in the labor composition. A bias resulting
from aggregation weakens the indices’ performance to denote wage levels.
Therefore, in order to monitor solely the wage developments, a wage index
measuring the value of a basket comprising well-defined components over time,
such as a price index, is needed. The Building Construction Cost Index, which is
released as a set of two indices for the construction sector, one for material costs
and the other for labor costs, is a good example to address this need. This index is
a price index that measures the changes in construction input costs by periods.
The labor cost index, which is constructed by aggregating the wages paid for
twelve labor items, is calculated by using 2005 weights. Average wage indices
displayed an increase with the crisis, whereas building labor costs exhibited a
decline (Chart 3). This outlook supports the idea that the increase in average
wages is driven by the change in the distribution of employees rather than the
wage developments.
Inflation Report 2011-III
63
Central Bank of the Republic of Turkey
An analysis of general trends in wage indicators demonstrates that average real
wages in industrial and construction sectors decreased over 2009 following the
crisis, before increasing again in 2010. In the services sector, which was relatively
less affected by global problems as it is less sensitive to external conditions, real
wages kept increasing at a limited rate in the crisis period and at a faster pace
since the second half of 2010.
Driven
by domestic demand, the robust trend of growth in Turkey since the
second quarter of 2009 highlights concerns over “overheating”. Real earnings and
labor cost indices show no evidence of an overheating economy. What is more,
wage developments should be examined together with the productivity
developments in this perspective.
When
productivity
outpaces/lags
Chart 6. Unit Wages
(Wage per Hour Worked/ Productivity)
(Real, 2008=100, Seasonally Adjusted)
Industrial
behind wages, unit wage costs may
decrease/increase. In an economy
110
with high productivity gains, rapid
100
growth can be sustained without
leading to inflationary pressures. An
analysis of unit wages calculated by
Short-Term Business Statistics indicates
that productivity increased across all
main types of business activity and
contributed to disinflation in the post-
Construction
Services*
115
105
95
90
85
80
75
I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I
2005
2006
2007
2008
2009
2010 2011
*In the services sector, real unit wages are calculated by
dividing total wage payments by turnover. In industrial and
construction sectors, total wage payments are divided by
output and CPI.
Source: TurkStat Short-Term Business Statistics, CBRT.
crisis period. The ongoing downtrend in unit wages as of the first quarter of 2011
indicates that wage developments do not imply overheating (Chart 6).
64
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Box
4.3
What the Economic Clock Says About Current Economic Activity
Identifying the current state of the economy and estimating its future course is
crucial for decision-making process of the economic agents. Leading indicators
constructed by macroeconomic data are among the most influential tools in the
economic literature in order to determine the turning points in the economy.
This
Box integrates the economic clock approach to coincident and leading
indicators index for the Turkish economy constructed by using the OECD-CACIS
method7, and thereby provides information about current economic activity and
its possible future course.
Economic
clock is used in order to analyze the state and the course of the
economy, where the horizontal axis shows the standardized level8 of the index
while the vertical axis denotes its month-on-month change (Figure 1). Different
states of the economy correspond to different quadrants in the economic clock.
In the first quadrant, the standardized level is below zero, indicating a month-onmonth increase. This may imply that the economy is coming out of the bottom. In
the second quadrant, the level is above the value implied by its long-term trend
and continues to rise month-on-month, pointing to an economic growth. In the
third quadrant, the level is above zero, but shows a monthly decline, indicating a
slowdown (below-average growth) in economic activity. Lastly, in the fourth
quadrant, both the level and the monthly growth rate are below zero, which
denotes that economic activity remains below the potential implied by its longterm trend and grows less than the average. As illustrated by the figure, the
economic clock is supposed to move clockwise.
7
Leading indicators are constructed by using the “Cyclical Analysis and Composite Indicators System (CACIS)” program that
facilitates the development of cyclical analysis and leading indicators by the OECD. For more information on the method used
by the OECD, please refer to http://www.oecd.org/dataoecd/37/42/42495745.pdf.
8
Standardized level is obtained by dividing the long-term average of x minus x (̅ ) to its standard deviation: ̅ ⁄
Inflation Report 2011-III
65
Central Bank of the Republic of Turkey
Figure 1. Economic Clock
In
constructing coincident and leading indicators, approximately 175 series
consisting
of
up-to-date,
economically
and
statistically
significant
price,
production, consumption, money-banking, foreign trade, balance of payments
and survey data that would serve as an indicator of economic activity were
analyzed. As an indicator for the general economic activity, industrial production
index was selected as the reference series. These series were classified as leading,
coincident and lagging by statistical analysis including their cross correlation with
industrial production and peak-trough analysis.
By
merging the different series presented in Table 1, several coincident and
leading indicator indices were constructed. Statistically best-performing indices of
coincident
and
leading
indicators
according
to
cross
correlation
and
peak/trough analysis are shown in Chart 1. The selected index of coincident
indicators is composed of capacity utilization rate, transport vehicles, electricity,
raw steel production and CNBC-e consumption index. The series constructing the
index of leading indicators are OECD’s leading indicators for member countries,
auction rates on GDBS (compound), 3-month ahead orders from BTS, Real Sector
Confidence Index and import quantity index for intermediate goods.
66
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Table 1. Classification of Indicators
Lagging Indicators
Coincident indicators
(PMI) Input inventory
CPI
CPI-based real effective
exchange rate (2003=100)
CPI Emerging Economiesbased real effective exchange
rate (2003=100)
CPI Advanced economiesbased real effective exchange
rate (2003=100)
(PMI) Input prices
(PMI) Output prices
(PMI) Delivery times
(PMI) Output Inventory
Use of credit cards
Total exports
Consumer loans
Consumption index
(credit/debit card
transactions index, real,
excluding food)
Consumption index
(credit/debit card
transactions index, real,
total)
Domestic loan utilization
Leading Indicators
(CNBC-e) Consumer Expectation, Tendency and
Confidence Indices
(PMI) Work backlogs
(PMI) Export orders
(PMI) Employment
(PMI) Quantity of purchases
(PMI) Production
SCA-H
(PMI) New orders
SCA-I
Foreign exchange basket
Electricity production
Raw steel production
Services prices
Total number of liquidated firms
Capacity utilization rate
Automobile production
Imports of capital goods
Production of commercial
vehicles
TEA Export figures
(CBRT) Expectations Survey Questions
(CBRT) Business Tendency Survey Questions
(CBRT) Reel Sector Confidence Index
(CBRT) Consumer Confidence Index
PMI Index
Export quantity index
Import quantity index
Imports of intermediate goods
Import quantity index for intermediate goods
Number of TIR transit permits
Number of TIR carnets
Total imports
(CNBC-e) Consumption Index
Chart 1. Cyclical Components of Coincident
and Leading Indicators
Coincident Indicators Index
103
Import quantity index for capital goods
Import quantity index for consumption goods
Domestic tax on goods and services (Nominal and
Real)
VAT on imports (Nominal and Real)
GDBS rate (Simple and compound)
Global industrial production index
ISE National Services Index
ISE National Industrial Index
ISE National 100 Index
ISE National Financial index
ISE National Technology Index
OECD Composite Leading Indicators
US Composite Leading Indicators
Balance of Payments, total commercial loans
Balance of Payments, commercial loans to banks
Domestic sales of white goods, exports and
production
Domestic sales and exports of automobiles
Domestic sales and exports of commercial vehicles
Number of new firms
PPI
PPI Energy
PPI Manufacturing
PPI Oil
Chart 2. Economic Clock for Economic Activity
I
Recovery
II
Growth
Leading Indicators Index
102
Standardized Level
101
100
99
98
-4
97
96
95
-2
0194
1294
1195
1096
0997
0898
0799
0600
0501
0402
0303
0204
0105
1205
1106
1007
0908
0809
0710
0611
94
Source: CBRT
Inflation Report 2011-III
0
2
May 2011
IV
Recession
January
2009
III
Slowdown
Monthly Change
Source: CBRT.
67
Central Bank of the Republic of Turkey
The
economic clock for the economic activity constructed by using industrial
production index cycle is shown in Chart 2. According to the economic clock, the
growth pace that started in June 2010 lost momentum in March 2011.
Chart 3. Economic Clock for Coincident
Indicators
-3
II
Growth
-2
-1
0
1
Monthly Change
Source: CBRT
Economic
-6
June
2011
January
2009
IV
Recession
2
I
Recovery
Standardized Level
Standardized Level
I
Recovery
Chart 4. Economic Clock for Leading Indicators
III
Slowdown
IV
Recession
II
Growth
-4
-2
0
January
2009
Monthly Change
2
June 2011
III
Slowdown
Source: CBRT.
clocks for coincident and leading indicators are shown in Charts 3
and 4, respectively. The economic clock constructed for leading indicators signals
signs of economic growth in February 2010, while the economic clock
constructed for coincident indicators points to August 2010 as the start of the
economic growth cycle, as expected by the methodology. Similarly, the
economic clock for leading indicators points to January 2011, whereas the
economic clock for coincident indicators indicates March 2011 as the start of the
slowdown in economic activity. The coincident indicators index, as expected,
monitors the changes in the economy contemporaneously, while leading
indicators index helps to foresee the developments in the economy.
Recent
observations as of June 2011 reveal that the economic clock for the
coincident indicators index points to a more pronounced deceleration in
economic activity in the second quarter. The economic clock constructed for the
leading indicators index exhibits a pattern of converging points in the third
quadrant in the recent months, which indicates that economic growth will remain
below average for a while.
68
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Box
Fixed Capital Growth Loss during the Recent Crises and Its
4.4
Impact on the Potential GDP
The question of overheating in the economy has recently been a top agenda
item. “Overheating” of an economy occurs when aggregate demand
remarkably exceeds the productive capacity and upward pressures on inflation
become more pronounced. For a sound analysis, a careful analysis of supply and
demand dynamics of the economy is required. In this context, a close scrutiny of
the dynamics regarding fixed capital investments (machinery and equipment
investments and construction investments), which are major contributors to
productive capacity, is significant in elucidating these discussions.
Past international experiences show that deep financial crises may bring about
persistent negative impacts on productive capacity.9 If a crisis leads to a
significant loss in productive capacity, the post-crisis recovery in demand would
pose risks to price stability by stretching the limits of the productive capacity and
require central banks to tighten their monetary policies. In this context, the impact
of the last global crisis on productive capacity is an agenda item of the CBRT as
well as many central banks around the world.
one of the significant impacts of crises on productive capacity is through slowing
down the capital growth.10 In economic crises, workplaces and manufacturing
facilities become idle, leading to a slowdown in existing or planned investments,
which has an adverse effect on the future productive capacity of the national
economy. For instance, vacant hotels in the crisis period cause existing or
planned hotel investments to lose pace, which in turn would reduce future
lodging capacity. This loss in capacity would not be significant during the crisis as
the operating hotels are also less occupied. However, as the economy rebounds,
occupancy rates may surge more rapidly due to the said capacity loss. Therefore,
this capacity loss may give way to upward pressures on accommodation prices
during economic recovery.
9
Furceri and Mourougene (2009), Cerra and Saxena (2008).
“Recessions typically have little effect on potential output beyond the direct effect of lower investment on capital
accumulation, and that effect tends to diminish in the long run when investment recovers to normal levels.” U.S. Congressional
Budget Office, The Budget And Economic Outlook: Fiscal Years 2010 To 2020, p.38. In addition to the capital effect, the global
crisis may have slowed down the potential growth by creating adverse and lasting effect on the financial system and the
labor markets in some countries (e.g. in the U.S.). However, these effects are believed to be relatively limited in the Turkish case.
10
Inflation Report 2011-III
69
Central Bank of the Republic of Turkey
Slowdown of investment during 2001 and 2009 crises in Turkey are illustrated in
Chart 1. Although investments fell drastically both during the 2001 crisis and the
recent one, the fall in the latter crisis is more limited and short-lived than in the
former one. The capital growth is shown in Chart 2. The annual average capital
growth, which was 5 percent in the 1987-2010 period, went remarkably below this
average during both crises; however, the fall in the 2001 crisis was more acute
and long-lasting. The amount of cumulative capital loss due to these belowaverage growth rates is illustrated separately for both crises in Chart 3 as a
function of the time passed since the onset of the crisis. Capital loss reached
nearly 11 percent in the 2001 crisis, but remained around 4 percent in the last
crisis.
By multiplying the capital losses in Chart 3 by 0.5, the estimated value for the
capital income share for Turkey, potential GDP effects for these losses can be
roughly estimated. This estimation shows that potential GDP loss reached 5.5
percent in the 2001 crisis (11%*0.5=5.5%), and hovered around 2 percent in the
last crisis (4%*0.5=2%).
How large and long-lasting the capital loss (and the resulting potential GDP loss)
can be is shown in Chart 3 for the 2001 crisis. The capital loss, which led to a GDP
loss of 5.5 percent in the 2001 crisis, was recovered only roughly by 2008. In
contrast, the potential GDP loss incurred by the capital loss in the last crisis led to a
potential GDP loss of only 2 percent and has already been recovered by around
40 percent so far on the back of the rapid recovery in investments. In other words,
the remaining potential GDP loss is around 1 percent as of mid-2011.
In short, the capital loss that occurred in the last crisis (and the resulting
potential
GDP loss) remained more limited than in the 2001 crisis and this loss was largely
compensated thanks to the rapid recovery in investments. This caused the
capacity loss in the economy to remain fairly limited in the last crisis, and played
an important role in preventing capacity utilization rates from reaching levels that
would create inflationary pressures (i.e. “overheating”), despite the robust
economic recovery following the crisis.
70
Inflation Report 2011-III
Central Bank of the Republic of Turkey
Chart 1. Fixed Investments
Chart 2. Fixed Capital Index Growth*
(Percent of GDP, Seasonally-Adjusted)
(Annual Percent Change, Seasonally-Adjusted)
26
10
24
8
22
6
20
4
18
2
0
16
-2
1998Q1
1998Q4
1999Q3
2000Q2
2001Q1
2001Q4
2002Q3
2003Q2
2004Q1
2004Q4
2005Q3
2006Q2
2007Q1
2007Q4
2008Q3
2009Q2
2010Q1
2010Q4
14
1987
1988
1989
1991
1992
1993
1994
1996
1997
1998
1999
2001
2002
2003
2004
2006
2007
2008
2009
2011
-4
* Figures for after 2011Q1 are CBRT estimates. The dashed line shows
the sampling average.
Source: Demiroğlu (2011), TurkStat, CBRT.
Source: TurkStat, CBRT.
Chart 3. Effects of 2001 and 2009 Crises on Capital Index*
(Deviation of Index Level from Average Growth Trend in Percent)
Percent
0.0
2009 crisis
2008Q2 = 0
-2.5
2008Q2=-1.7
2011Q2=-2.6
-4%
-5.0
2001 crisis
2000Q3 = 0
-7.5
-10.0
-11%
Quarter
-12.5
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
* The horizontal axis shows the quarter following the crisis.
2011 figures are estimates. Capital share is assumed to be 50 percent.
Source: Demiroğlu (2011), TurkStat, CBRT.
REFERENCES
Cerra, V. and S. Saxena, 2008, Growth Dynamics: The Myth of Economic
Recovery, American Economic Review, 98(1): 439-57.
Demiroğlu, U., 2011, Losses in Fixed Capital Growth of Turkey in Recent Crises and
the Impact of these Losses on the Potential GDP, CBRT Economic Note
(forthcoming).
Furceri, D. and A. Mourougane, 2009, The Effect Of Financial Crises On Potential
Output: New Empirical Evidence From OECD Countries, OECD Economics
Department Working Paper No. 669.
Inflation Report 2011-III
71
Central Bank of the Republic of Turkey
72
Inflation Report 2011-III