India’s accelerated economic growth in recent years has been a focus of significant policy discussion and analysis. The services sector has played a pivotal role in this acceleration. Growth in services picked up in the 80’s and accelerated in the 90’s.  Since then, it has become a dominant contributor to economic growth. According to Rakesh Mohan (2008) “...it is the continuing and consistent acceleration in growth in services over the decades, that had earlier been ignored, that really accounts for the continuous acceleration in overall GDP growth*”. With the increasing use of computers and telecommunication in business transactions, the domain of the services sector is growing bigger day by day. The software industry, and particularly its share in external transactions, has grown at a rapid pace during the last decade.

Services sector growth has been higher than the growth in other sectors such as agriculture and manufacturing sectors. In fact, India’s rising per capita GDP is associated with an increase in services. As the economy has grown, the sectoral share given up by the agricultural sector has shifted more to the services sector than to industry. The services sector’s share in the gross domestic products has increased from 40.3 per cent in the 80s to 45.7 per cent in the 90’s and further to 53.1 per cent in 2000-07. The services sector grew at an annual average of 6.8 per cent in the 80s, 7.6 per cent in the 90s and 9.3 per cent in 2000-07. The years 1995-96 and 2006-07 recorded the high year-to-year growth rate of over 10 per cent in the services sector. The entire decline in the share of agriculture sector in GDP has been picked up by the services sector while manufacturing sector’s share has remained more or less the same over the years (Table-1).

The services sector in India includes trade, hotels and restaurants, railways, transportation by other means, storage and communication, financing, insurance, ownership of dwellings, real estate, business and legal services, public administration and defence and community and personal services. Table-1 below shows that that the prime movers of the growth in services are hotels and restaurants, communication and banking and business services (computer related services, renting of machinery, accounting and research development) with recorded growth rates above 10 per cent. Communication services needs special mention because of the phenomenal rise in its growth rate from 5.6 per cent in the 1980s to 15.5 per cent in the 1990s and further to 25.8 per cent in 2000-07.  The figures also show that in the case of business services and hotels and restaurants, there was a significant acceleration in growth in the 1990s and this was maintained in the 2000-2007 period. Other sectors which have shown a significant rise in growth rates include the real estate and other services (community, personal services). The insurance sector was relatively slower to take off, with a sharp step-up only in 2000-2007. However, the growth rate of services like ownership of dwellings, public administration and defence fell drastically over the decades.

      Table-1: Growth rates and shares of the services sector components

*: Compound growth rates are calculated by estimating semi-log trend equations except for 2008-09
#: Revised estimates;                             Source: NAS, RBI

Complementarity between the services and other sectors:
There is a vast literature in economics supporting complementarity between the industry and services sector. Fast growing services like trade and transport, which are also called producer services, reflect the complementarity between industry and services sectors. The regression analysis below was done to test whether such a complementarity exists and the extent of such complementarity between the services and the two sectors of manufacturing and infrastructure. We have also looked at the impact of public investment in services and infrastructure on private investment in services.

Since, all the variables are stationary in log form; a log-linear function has been used to estimate in the equations. In all the equations below, ‘Ln’ indicates natural logarithm. The ‘T-statistic’ for each variable is given in the brackets below the coefficients. ‘AR’ term is used in each equation to correct for serial correlation. We have used data for the period from 1981-82 to 2007-08 for the OLS estimation. Since, the equations are estimated in log-liner form, the coefficients can be considered as partial elasticities.

Complementarity between manufacturing and services sector:
The equation below describes the complementarity between industry and services sectors. Real output in services is estimated as a function of real net capital stock in services and real output in manufacturing.

1. Ln YSRR = 4.42 + 0.59 Ln KSRR + 0.75 Ln YMNR + 0.57 AR (1)
            (6.4)     (4.4)                       (6.8)                             (2.9)
           
                        `R2 = 0.99                   DW = 1.5
Variables:
KSRR: Real net capital stock in services
YMNR: Real output in manufacturing
YSRR: Real output in services sector

As we observe from the equation that real output in manufacturing sector has a strong, positive effect on services sector output with an elasticity of 0.75 i.e. for every 10 per cent increase in manufacturing output, services sector output will increase by 7.5 per cent. As expected, net capital stock in services affects services output positively. 

Complementarity between infrastructure and services sector:
The importance of infrastructure in facilitating services is well recognised. Having regard to the current level of service’s share in the total output, strengthening of the infrastructure would have a favourable impact on the growth of service sector. In this context it will be interesting to observe the complementarity between infrastructure and services output. The following equation shows the relationship between real output in services and infrastructure, which includes electricity, gas, water supply and construction.

2. Ln YSRR = 2.33 + 0.49 Ln KSRR + 0.74 Ln YINFR + 0.66 AR (1)
             (2.1)     (3.1)                      (6.6)                             (4.5)
           
                        `R2 = 0.99                   DW = 2.03
Variables:
KSRR: Real net capital stock in services
YINFR: Real output in infrastructure
YSRR: Real output in services sector

The real net capital stock has a positive effect on services output as expected. Real output in infrastructure has a positive impact on services output with an elasticity of 0.74.  There seems to be a significant complementarity between services and infrastructure sectors.

Complementarity between private investment in the services sector and public investment in infrastructure:

It will also be interesting to check the complementarity between private investment in the services sector and public investment in infrastructure and the crowding in/out effects between private and public investments in the services sector. The equation below estimates real private investment in the services sector as a function of real output in services sector, real public investment in services sector as well as in infrastructure and the real interest rate.

3. Ln PVTCFSRR = -11.39 + 1.44 Ln YSRR + 0.13 Ln PCFSRR + 0.22 Ln PCFINFR
                         (-4.0)     (4.1)                      (0.3)                                   (0.6)

                                                                        - 0.03 Ln (PLR-INFL) - 0.18 AR (1)
   (-1.7)                             (-0.65)
                                                                                                                                     
                        `R2 = 0.96                   DW = 1.96
Variables:
INFL: Rate of inflation
PCFINFR: Real public investment in infrastructure
PCFSRR: Real public investment in services
PLR: Prime lending rate
PVTCFSRR: Real private investment in services
YSRR: Real output in services    

The real output has a significant positive effect on private investment in the services sector following acceleration principle. Public investment in both services and infrastructure has a positive impact on private investment in services though not significant.

Service sector growth and Inflation:
It is argued that services-led growth might lead to inflation as the services sector by definition is not a commodity producing sector (Bhattacharya and Mitra 1990; IMF 2006). The rationale behind this argument is that there will be a demand-supply gap created because of the increase in output without a simultaneous change in consumer goods produced. This gap might have an inflationary effect on the economy. A faster growth in services may also lead to a divergence between the inflation rates in traded and non-traded goods sectors. To get a deeper insight into this issue, we did the following empirical analysis.  It estimates the impact of the share of the services sector output in GDP on the price level (as reflected in the wholesale price index).

The wholesale price index is estimated as a function of the share of the services sector output in total GDP, money supply, lagged wholesale price index and a dummy variable for monetary policy. The dummy was introduced in order to capture the effectiveness of monetary policy in containing inflation in the economy. The dummy variable takes the value ‘1’ when the inflation is below 5 per cent and ‘0’ otherwise.  Here also we estimated the equation in a log-linear form. ‘Ln’ indicates natural logarithm. ‘T-statistic’ for each variable is given in the brackets below the coefficients. ‘AR’ term is used in the equation to correct for the serial correlation. We used the period from 1981-82 to 2007-08 for OLS estimation.

4. Ln P = -0.04 - 0.85 SHYSR + 0.04 Ln M3 - 0.03 DUMP + 0.96 Ln P-1                         
    (-0.1)    (2.6)                   (1.2)                           (-4.3)                            (16.4)
                   
                        `R2 = 0.99                   DW = 1.5

Variables:
DUMP: Dummy variable for monetary policy
M3: Money supply
P: Wholesale price index
SHYSR: share of the services sector output in total GDP (nominal)

The above regression analysis clearly shows that an increase in the share of the services sector in total output does not have a positive impact on the wholesale price index. Hence, a rise in the share of the services sector in GDP with an effective monetary policy will control the inflation as indicated through the sign and level of significance of their respective coefficients.

Concluding observations: