SPECIAL ARTICLE An Index to Assess the Stance of Monetary Policy in India in the Post-Reform Period
Amaresh Samantaraya
The Reserve Bank of India has formally adopted the “multiple indicator approach” in the conduct of monetary policy since April 1998. During this period, sole reliance on traditional indicators of monetary aggregates or policy rates is not adequate to reflect the stance of monetary policy. This paper develops a monetary policy index by synthesising the extracted signals from the policy documents and quantitative information embedded in key indicators. The MPI so constructed was used to assess the impact of monetary policy on macroeconomic variables such as interest rates, bank credit, inflation, and output growth during the post-reform period. It was observed that while monetary policy has an instant influence on interest rates, the impact on inflation and output was realised with a lag of around 6 to 18 months.
The author is grateful to B Kamaiah and an anonymous referee of this journal for very useful suggestions. Needless to mention, the responsibility for errors, if any, lies only with the author. The paper expresses the author’s own views and does not necessarily reflect those of the Reserve Bank of India.
Amaresh Samantaraya (asamantaraya@rbi.org.in) is with the Reserve Bank of India.
M
In general, monetary aggregates, policy rates/central bank target rates are used to capture the stance of monetary policy. In r ecent years, the issues related to stability of the money demand function and money multiplier has triggered an exploration of alternate indicators to monetary aggregates for evaluation of the effectiveness of monetary policy. Moreover, given the fact that both monetary aggregates and interest rates are prone to fluctuations independent of monetary policy stance, Romer and Romer (2004) had introduced an alternative “narrative approach” d erived from the policy statements of the monetary authority. In Indian context, a recent study by Bhattacharyya and Ray (2007) attempted to empirically examine the impact of monetary policy shocks on prices and output using a measure of the monetary policy stance derived from the detailed reading of monetary p olicy announcements during the period 1973 to 1998.
The present paper attempts to further extend the approach to the latest period by constructing a monthly monetary policy i ndex (MPI) for the post-reform period1 in India, i e, April 1996 to June 2008. Using an index by combining a variety of indicators to reflect monetary policy stance in India has assumed greater relevance during this period with the Reserve Bank of India (RBI) adopting “multiple indicator approach” in conduct of monetary policy since April 1998. In this approach, a host of indicators i ncluding interest rates, exchange rate, monetary and banking aggregates and prices are pooled together for drawing policy perspectives. In the context of de-emphasising exclusive reliance on monetary aggregates with this approach, RBI’s
May 16, 2009 vol xliv no 20
SPECIAL ARTICLE
S tatement on Monetary and Credit Policy for 1999-2000 observed that,
the objective is to widen the range of variables that could be taken into account for monetary policy purposes rather than rely solely on a single instrument variable, such as, growth in broad money (M3). The experience of 1998-99 shows that money supply movements alone could not have provided adequate guidance for monetary policy initiatives during the year.
Mohan (2005) also indicated some incoherent association between growth in monetary aggregates and inflation outcomes from cross-country experience. The role of interest rate as a proxy for monetary policy stance during this period has also several limitations. First, the period of study witnessed transition from the bank rate as the signalling device to repo/reverse repo rates under the Liquidity Adjustment Facility (LAF). Second, unprecedented international capital inflows d epressed interest rates even though tight money policy was adopted in several occasions.
Thus, considering the limitations of sole reliance on traditional indicators of monetary aggregates or policy rates to reflect the stance of monetary policy adequately during 1996-2008, this p aper proposes a MPI blending different indicators of monetary policy. In addition to the subjective information on the stance of monetary policy as revealed by pronouncements made in various policy documents, the MPI developed here also takes into account additional quantitative information embedded in key monetary policy indicators and thus comprehends a broad information set reflecting monetary policy more appropriately. We also presume that this approach may help in reducing some element of subjectivity inherent in the MPI proposed by Bhattacharyya and Ray (2007). The MPI so constructed was used to empirically assess the impact of monetary policy on key macroeconomic variables such as interest rates, bank credit, output and inflation in India. Data used in this paper are obtained from the Handbook of Statistics on Indian Economy, 2007-08 published by the RBI.
The organisation of the paper is as follows: Section 1 highlights the background and stylised facts about monetary policy in India during the post-reform period. Section 2 describes the construction of the MPI for assessing monetary policy stance. Empirical evidences on the movement of the MPI vis-à-vis key macroeconomic variables are presented in Section 3, and Section 4 summarises the conclusions.
1 Background and Stylised Facts
1.1 Monetary Policy Framework in India
Under the Reserve Bank of India Act, 1934, the RBI is entrusted with the responsibility “to regulate the issue of Bank Notes and the keeping of reserves with a view to securing monetary stability in India and generally to operate the currency and credit system of the country to its advantage”. Accordingly, the main objectives of monetary policy in India have evolved as maintaining price stability and ensuring adequate flow of credit to the productive sectors to support economic growth. In the recent period, considerations of financial stability have assumed added importance in view of increasing openness of the Indian economy (Reddy 2007; Mohan 2007).
Economic & Political Weekly
EPW
The operating framework of monetary policy in India has undergone significant transformation during the past two decades in consonance with evolving changes in the structure of the economy and institutional developments. Prior to the mid-1980s, “credit planning” ensured flow of bank credit to the relevant s ectors as per the national priorities. With adoption of “monetary targeting with feedback” since the mid-1980s, broad money (M3) emerged as the nominal anchor based on the premise of a stable relationship between money, output and prices. In the late 1990s, ongoing financial openness and sweeping changes in the financial sector reoriented the role of interest rates vis-à-vis the quantity variables. It was felt that in the evolving situation, while money still acts as an important indicator, information pertaining to other monetary and financial indicators should also be taken into account while formulating monetary policy. Since April 1998, the RBI has formally adopted a “multiple indicator” approach in which information on interest rates, monetary a ggregates, credit, capital flows, inflation, exchange rate, etc, are pooled together for drawing policy perspectives. During this p eriod, as discussed earlier, sole reliance on traditional indicators of monetary aggregates or interest rates is not adequate to reflect the stance of monetary policy, and hence, it is important to c arefully comprehend enunciations in policy statements juxtaposed with information on relevant policy indicators to decipher the complete picture.
1.2 Monetary Policy Instruments
In the evolving monetary policy framework, while greater reliance was placed on indirect instruments, there was no hesitation in taking recourse to direct instruments if circumstances so w arranted. The main instruments that the RBI used to implement monetary policy during our period of study include: t Open Market Operations (OMO): OMO include both outright transactions in government securities as well as repo/reverse repo operations under the LAF. Repo/reverse repo rates under LAF (since April 2000) modulate day to day liquidity in the system and steers desired trajectory of money market interest rates, thus emerging as the main instruments for interest rate signalling. R efinance facilities for the banks at present are provided at repo rate under LAF. t Bank Rate: The Bank Rate was reactivated as the central bank signalling rate since April 1997 by linking it to various rates of r efinance. With emergence of repo/reverse repo rates under LAF as the effective signalling rates, refinance facilities are gradually delinked from the Bank Rate. t Market Stabilisation Scheme (MSS): The MSS was introduced in March 2004 as part of special sterilisation operations in the wake of very large and continuous capital inflows and the need for modulating surplus liquidity conditions. While these issuances do not provide budgetary support, interest costs are borne by the fisc. As far as government securities market is concerned, MSS securities are also traded in the secondary market, at par with the other government stock. t Cash Reserve Ratio (CRR): In the post-reform period, the m edium-term policy was to gradually reduce CRR to its statutory minimum. But, in response to unprecedented surge in foreign
SPECIAL ARTICLE
| Table 1: Monetary Policy Index for India | to tightening mode since late 2004. Reverse repo/repo rate under |
| Month-Year MPI Month-Year MPI Month-Year MPI Month-Year MPI | LAF were gradually raised from 4.5%/6% in the mid-2004 to 6.% |
| Apr-96 0.25 May-99 0.83 Jun-02 0.8 Jul-05 0.27 | (July 2006)/8.50% (June 2008). CRR was raised gradually from |
| May-96 0.24 Jun-99 0.83 Jul-02 0.81 Aug-05 0.27 | 5% prior to December 2006 to 8.25% by May 2008. Moreover, to |
| Jun-96 0.27 Jul-99 0.83 Aug-02 0.81 Sep-05 0.36 Jul-96 0.33 Aug-99 0.79 Sep-02 0.81 Oct-05 0.32 | arrest the unprecedented credit growth provisioning norms on standard assets were raised from 0.25% to 0.4% in October 2005. |
| Aug-96 0.3 Sep-99 0.78 Oct-02 0.82 Nov-05 0.33 Sep-96 0.28 Oct-99 0.76 Nov-02 0.81 Dec-05 0.34 Oct-96 0.5 Nov-99 0.79 Dec-02 0.8 Jan-06 0.28 | With excessive credit growth in certain sectors such as retail, commercial real estate, credit card, etc, and the need to sensitise |
| Nov-96 0.55 Dec-99 0.84 Jan-03 0.81 Feb-06 0.3 | the banks about the implications for credit quality, provisioning |
| Dec-96 0.53 Jan-00 0.79 Feb-03 0.8 Mar-06 0.4 | on standard assets on those segments were raised to 1% in April |
| Jan-97 0.59 Feb-00 0.78 Mar-03 0.77 Apr-06 0.61 | 2006 and further to 2% since January 2007. |
| Feb-97 0.58 Mar-00 0.74 Apr-03 0.79 May-06 0.6 | |
| Mar-97 0.57 Apr-00 0.79 May-03 0.71 Jun-06 0.59 | 2 Construction of the MPI |
| Apr-97 0.35 May-00 0.78 Jun-03 0.73 Jul-06 0.37 | As discussed, against the backdrop of limitations of traditional |
| May-97 0.32 Jun-00 0.76 Jul-03 0.73 Aug-06 0.38 | indicators of monetary policy such as monetary aggregates or |
| Jun-97 0.33 Jul-00 0.27 Aug-03 0.72 Sep-06 0.36 Jul-97 0.33 Aug-00 0.21 Sep-03 0.73 Oct-06 0.34 Aug-97 0.31 Sep-00 0.23 Oct-03 0.74 Nov-06 0.36 Sep-97 0.3 Oct-00 0.5 Nov-03 0.73 Dec-06 0.34 Oct-97 0.33 Nov-00 0.54 Dec-03 0.75 Jan-07 0.38 | policy rates to proxy monetary policy stance in the recent years, the present study proposed to develop a more comprehensive MPI to fill the gap. The MPI constructed (Table 1) is based not only on monetary policy signals derived from the governor’s policy state |
| Nov-97 0.33 Dec-00 0.53 Jan-04 0.74 Feb-07 0.41 | ments,3 but also incorporates information on growth of broad |
| Dec-97 0.31 Jan-01 0.53 Feb-04 0.79 Mar-07 0.33 | money (M3) and informal operating target of “call money rate”. |
| Jan-98 0.1 Feb-01 0.53 Mar-04 0.83 Apr-07 0.36 | In the “multiple indicator approach” followed by the RBI mone |
| Feb-98 0.29 Mar-01 0.55 Apr-04 0.82 May-07 0.36 | tary aggregate such as “M3” remains as a key indicator. On the |
| Mar-98 0.32 Apr-01 0.81 May-04 0.8 Jun-07 0.45 | other hand, in the liquidity management operations, the RBI aims |
| Apr-98 0.59 May-01 0.82 Jun-04 0.8 Jul-07 0.46 | at maintaining the inter-bank “call money rate” within the repo |
| May-98 0.59 Jun-01 0.82 Jul-04 0.8 Aug-07 0.39 | reverse repo corridor under the LAF. Thus, information on M3 |
| Jun-98 0.59 Jul-01 0.82 Aug-04 0.81 Sep-07 0.4 Jul-98 0.61 Aug-01 0.83 Sep-04 0.78 Oct-07 0.43 Aug-98 0.64 Sep-01 0.81 Oct-04 0.53 Nov-07 0.43 Sep-98 0.63 Oct-01 0.79 Nov-04 0.51 Dec-07 0.42 Oct-98 0.64 Nov-01 0.77 Dec-04 0.5 Jan-08 0.45 | growth and call money rate juxtaposed with stance as enunciated in the governor’s policy statements would suitably reflect overall monetary policy stance. Our approach is to construct a MPI for I ndia similar to that used by UNDP for computation of some well |
| Nov-98 0.62 Dec-01 0.75 Jan-05 0.54 Feb-08 0.43 | known development indices such as the human development |
| Dec-98 0.61 Jan-02 0.76 Feb-05 0.5 Mar-08 0.39 | i ndex (HDI), gender-related development index (GDI) and so on.4 |
| Jan-99 0.61 Feb-02 0.76 Mar-05 0.49 Apr-08 0.40 | The MPI is constructed summarising three sub-indices. Each sub |
| Feb-99 0.61 Mar-02 0.75 Apr-05 0.26 May-08 0.43 | index is computed by the following formula: |
| Mar-99 0.6 Apr-02 0.76 May-05 0.27 Jun-08 0.38 | |
| Apr-99 0.84 May-02 0.82 Jun-05 0.26 | SIi = (Ai – xi)/(Xi – x i) …(1) |
| capital inflows, CRR was reactivated since December 2006 as a | where, Ai = actual value of the indicator ‘i’, xi = minimum value |
| monetary policy instrument in the sterilisation process. | of the indicator ‘i’, Xi = maximum value of the indicator ‘i’ |
| Formula (1) ensures that the sub-index lies between ‘0’ and | |
| 1.3 Stylised Facts | ‘1’. Starting with the sub-index derived from RBI governor’s |
| Broad characteristic of monetary policy in India during April | statements, denoted as SI1 – an interpretation of expansionary |
| 1996 to June 2008 can be summarised with two stylised facts:2 | policy was assigned a score of ‘1’, while those of contractionary |
| t Accommodating monetary policy began in the second half of | and neutral policy stances were assigned scores of ‘–1’ and ‘0’, |
| 1996-97 to support credit expansion in the face of economic slow | respectively. SI1 is obtained using Formula (1), where x1 = –1, |
| down and largely continued until the mid-2004. The Bank Rate | X1 = 1 and A1 for a particular month takes values as per the |
| was gradually reduced from 11% in April 1997 to 6% by April | i nterpretation of expansionary/neutral/contractionary as |
| 2003, while CRR was brought down in a phased manner from | d iscussed above. |
| 13.5% in April 1996 to 4.5% in June 2003. However, monetary | The second sub-index, denoted as SI2 is based on year-on-year |
| policy had resorted to temporary tightening mode during | broad money (M3) growth for a particular month. For computa |
| 1997-98 to guard against the repercussions of east Asian crisis | tion of SI2, we have used x2 = 11.18 (lowest M3 growth rate during |
| and again in the mid-2000 responding to developments in the | 1996-2008) and X2 = 23.65 (highest) and A2 as equal to monthly |
| foreign exchange market. | actual M3 growth rate. |
| t In the face of apparent signs of overheating with economic | The third sub-index, denoted as SI3 is based on the “call money |
| growth surpassing 8%, credit growth at unprecedented high of | rate”. For computation of SI3, x3 = 0.73 (lowest rate for 1996 |
| around 30% and elevated assets prices, monetary policy switched | 2008: July 2007) and X3 = 23.65 (highest: January 1998) and A3 |
| 48 | may 16, 2009 vol xliv no 20 Economic & Political Weekly |
EPW| SPECIAL ARTICLE | |||
|---|---|---|---|
| Graph 1: MPI and Interest Rates 0.8 | MPI | 16.0 12.0 | income, spendings on education and health, life expectancy, working abilities, etc. Despite this, HDI is the most popular index to measure human developments. |
3 MPI to Assess the Impact of Monetary Policy in India
5-yr GSEC
Credit Growth in % Interest rates in %
In order to analyse the impact of monetary policy on various
macro economic indicators, movement of MPI vis-à-vis interest
rates, bank credit, inflation, industrial growth are graphically
8.0
MPI
91-TBs
4.0
d epicted in Graphs 1 to 4. It can be observed from Graph 1 that along with expansionary monetary policy during 1998-2004,
0.0
Apr-96 Sep-97 Mar-99 Sep-2000 Mar-02 Sep-03 Mar-05 Sep-06 Jun-08 i nterest rates (91-day Treasury Bills rate (91-TB) and yield on g overnment securities of 5-year residual maturity (5-yr GSEC))
Graph 2: MPI and Bank Credit
e xhibited a downward trend, while with adoption of tight mone
40.0
tary policy since late 2004, interest rates firmed up gradually. T emporary monetary tightening during 2000 also witnessed
MPI 30.0
t emporary shooting up of the interest rates. There seems very
close and i nstant negative association between the MPI and i nterest
rate movements.
We have used three-month moving averages of year-on-year
20.0
MPI
0.4
Credit
10.0
growth rates of non-food credit of scheduled commercial banks,
0.0 Apr-96 Sep-97 Mar-99 Sep-2000 Mar-02 Sep-03 Mar-05 Sep-06 Jun-08
0
takes values equal to call money rate for a particular month. Combining the three sub-indices, MPI is computed as:
MPI = 1/2 (SI1) + 1/4 (SI2) + 1/4 (1 – SI3) …(2)
By construction, the value of MPI would lie between ‘0’ and ‘1’. A higher value of MPI indicates monetary expansion, while lower value indicates tight money policy. Closer the value to unity, higher is the degree of expansion and closer the value to zero, higher the degree of contraction. Thus, a downward movement of MPI in the scale indicates withdrawal of accommodation and an upward movement represents easy money policy.
From the movement of MPI (Graph 1) it can be observed that the RBI broadly followed easy money policy since late 1996 up to index of industrial production (IIP) and wholesale price index (WPI) as proxy for bank credit, industrial production and inflation. During our period of study, bank credit growth (CREDIT) roughly witnessed a rising trend from late 1990s up to late 2005 (Graph 2). This was preceded by the expansionary monetary policy until mid-2004 as discussed earlier. Similarly, moderation of the credit growth since 2006 is also preceded by monetary tightening as reflected in the movement of MPI. Even upward/downward cyclical swings in the bank credit growth within the broad trend is also marked to be preceded by expansionary/concretionary monetary policy with certain time lag.
Graph 3 provides indications of some close association b etween inflation and monetary policy. Every upward/downward
Graph 3: MPI and Inflation (WPI)
8.0
0.8
6.0
early 2004 with temporary tightening during 1997-98 and mid
0.6
2000. This is consistent with the empirical facts discussed above.
Inflation in % Inflation in % MPI
The MPI also captured tight money policy since late 2004 in the
0.4
INF
face of apparent signs of overheating.
If one construes policy pronouncements as the cause and variables like M3 growth and call money rate as the effect, then the MPI developed in this paper is suspected to suffer from circularity of causation.5 However, it may be noted that despite very close link, these indicators individually reflect monetary policy stance in someway or other. The motive for combining them was driven by the fact none of them, taken exclusively, do not reflect monetary policy stance properly. In a similar context, it may be noted that UNDP’s HDI also combines three basic dimensions of human
2.0
0.2
MPI 0
0.0
Apr-96 Sep-97 Mar-99 Sep-2000 Mar-02 Sep-03 Mar-05 Sep-06 Jun-08
Graph 4: MPI and Industrial Growth
0.8
12.0
0.6
8.0
developments, viz,(i) a long and healthy life, (ii) knowledge, and
MPI
0.4
(iii) a decent standard of living. A long and healthy life is repre
4.0
sented by life expectancy at birth, knowledge is represented by
IIP
0.2
MPI
gross enrolment ratio, and the standard of living is represented
by the GDP per capita. Amongst the components of HDI also there
0
0.0
could be circularity of causation given the interlinkages amongst Apr-96 Sep-97 Mar-99 Sep-2000 Mar-02 Sep-03 Mar-05 Sep-06 Jun-08
SPECIAL ARTICLE
swing in inflation (INF) is preceded by similar movements in the policy has an instant influence on interest rates, while it also policy stance. The systematic co-movements was of course with e xerts the expected impact on bank credit, industrial production certain lag. For example, corresponding to monetary tightening and inflation with some time lag. during the periods 1997-98, mid-2000 and 2004-08, downward movement in inflation is broadly discernible during 1999, 4 Conclusions 2001-02 and 2005-08. On the other hand, corresponding to ac-In view of the apparent inadequacy of traditional monetary polcommodating monetary policy during the 1996-97, 1998-99 to icy indicators to capture monetary policy stance appropriately for 2001-04, inflation exhibited an upward movement during the the post-reform period in India, the current paper suggested a p eriods 1997-98, 1999-2000 and 2002-05. Even temporary MPI as a potential alternative. The MPI constructed is a summary u pward movement in inflation of late was also preceded by less measure encompassing traditional policy indicators like monecontractionary monetary policy, as indicated by mild upward tary aggregates and overnight money market rate as also used movement in MPI. The lag of policy effect varied from around six policy signals from documented pronouncements. It was used for to 18 months. a preliminary assessment of the influence of monetary policy on
Some association between monetary policy and industrial pro-important macroeconomic variables relevant for monetary duction can also be discernable from Graph 4. Following expan-p olicy in India. It was observed that monetary policy has an sionary monetary policy in the first phase of our period of study, i mmediate impact on interest rates, while that on bank the industrial growth rate as measured by the IIP witnessed an credit, industrial production and inflation was realised with some upward movement in since 2001, while the mild moderation of time lag. However, the index developed in the paper need to be late is also preceded by monetary tightening. Thus, using our MPI used in formal econometric analysis to confirm the monetary to capture the monetary policy stance, we observed that m onetary policy transmission.
Notes
Reform initiatives towards greater monetary policy autonomy and facilitating its market orientation in India which began in early 1990s, found fuller expression in the post-1996-97 period. In the sphere of monetary-fiscal coordination, a landmark institutional reform enabling greater monetary policy autonomy was phasing out of issuance of ad hoc. Treasury Bills since April 1997 to curtail “automatic monetisation”. Reforms in the banking sector include freedom to determine lending rates for credit limit of over Rs 2 lakh since October 1994 and term deposit rates of maturity over 1 year since July 1996. With a view to promote the interest rate channel of monetary transmission, the bank rate was reactivated since April 1997 by linking it to various refinance rates. The SLR and CRR were r educed in a phased manner to rationalise statutory pre-emption of resources of the banks. The SLR was reduced to 25% (statutory minimum until June 2006) since October 1997 and the CRR witnessed gradual downward revision with a mediumterm objective of reducing it to the statutory minimum. In the government securities market, p rimary dealers (PDs) became operational since 1996 a ctivating secondary market trading, liquidity and turnover.
2 This excludes further monetary tightening with rising inflation and inflation expectation and subsequent change of direction with monetary accommodation in the face of spillovers of recent global financial turbulence on the Indian financial markets during July-October 2008 which is outside our period of study.
3 Policy documents referred to are annual policy statements of the governor and mid-term/quarterly reviews, at present and similar policy statements by the governor enunciating the Reserve Bank’s annual monetary and credit policy, previously and related mid-term reviews.
4 For details see UNDP’s Human Development R eports available from the UNDP web site (www. undp.org).
5 I am grateful to the anonymous referee to point out this issue.
References
Bhattacharyya, Indranil and Partha Ray (2007): “How Do We Assess Monetary Policy Stance? Characterisation of a Narrative Monetary Measure for India”, Economic & Political Weekly, 31 March, pp 1201-10.
Mohan, Rakesh (2005): “Some Apparent Puzzles for Contemporary Monetary Policy”, RBI Bulletin, December, pp 1109-22.
– (2007): “India’s Financial Sector Reforms: Fostering Growth While Containing Risks”, RBI Bulletin, December, pp 2199-2226.
Rangarajan, C (1997): “Role of Monetary Policy”, E conomic & Political Weekly, Vol XXXII, No 52, 27 December, pp 3325-28.
RBI (2003): Report of the Internal Group on Liquidity Adjustment Facility.
Reddy, Y V (2007): “Evolving Role of the Reserve Bank of India: Recent Developments”, RBI Bulletin, D ecember, pp 2199-2226.
Romer C D and D H Romer (2004): “A New Measure of Monetary Shocks: Derivation and Implications”, American Economic Review, Vol 94, No 4, pp 1055-84.
Samantaraya, Amaresh (2003): “Transmission Mechanism and Operating Procedure of Monetary P olicy in India: An Econometric Analysis”, PhD thesis submitted to the University of Hyderabad, Hyderabad.
– (2008): “Monetary Policy of the Central Bank: Simplifying the Mystique”, CAB Calling, Vol 32, No 2, April-June, pp 17-24.
REVIEW OF WOMEN’S STUDIES
April 25, 2009
Women’s Citizenship and the Private-Public Dichotomy – Maithreyi Krishnaraj Gendered Citizenship and Women’s Movement –Anurekha Chari Girl Abroad: The Private and the Public in Jab We Met... – Shoba Venkatesh Ghosh Exploring Gender, Hindutva and Seva – Swati Dyahadroy Maya Machhindra and Amar Jyoti: Reaffirmation of the Normative – Vaishali Diwakar Reading Devadasi Practice through Popular Marathi Literature – Anagha Tambe
For copies write to Circulation Manager
Economic and Political Weekly
320-321, A to Z Industrial Estate, Ganpatrao Kadam Marg, Lower Parel, Mumbai 400 013. email: circulation@epw.in
may 16, 2009 vol xliv no 20
