Financial Intermediation and Monetary Policy Transmission in EMEs: What has Changed Since the 2008 Crisis?

Abstract

In contrast to the benign neglect of the financial system in traditional monetary models, there has been growing evidence in recent years that the size and the structure of financial intermediation play a critical role in the transmission of monetary policy. This paper reviews the implications of three key post-2008 crisis developments in financial intermediation—the role of banks, the globalization of debt markets and the sustained decline in global long-term interest rates—for various transmission channels of monetary policy in EMEs. The paper argues that the globalization of debt markets means that monetary policy can no longer be conducted through the short-term interest rate alone. This raises questions about the appropriate instruments to be used for economic stabilization in this new environment.

Views expressed in this paper are those of the authors and not necessarily the views of the Bank for International Settlements, University of Basel or the Reserve Bank of India.

1 Introduction

The purpose of this chapter is to review what has changed to the monetary transmission mechanisms in emerging market economies (EMEs) since a similar review on this subject appeared in BIS (2008) and Mohanty and Turner (2008). A key finding then was that the introduction of inflation targeting by many EMEs in the 1990s, together with reforms to abolish interest rate controls, strengthen central bank credibility, and develop local bond markets, marked a major turning point for monetary policy in many countries. These reforms not only helped to reduce the earlier constraints on monetary policy stemming from a high degree of fiscal dominance and liability dollarization but also increased the role of interest rate and exchange rate in monetary policy transmission, leading to an environment of low and stable inflation.

However, the past decade has seen major changes to financial intermediation in EMEs, accompanied by rapid changes in external monetary environment following the global financial crisis in 2008. How have these developments affected the monetary transmission mechanisms in EMEs? Has the earlier assessment changed? And how have central banks confronted many recent changes to the external environment? Our objective here is to explore some of these questions in a fairly selective manner, drawing on a large, though still developing, post-crisis literature on monetary policy in EMEs.

Understanding how central banks’ instruments work has major implications for the stance of monetary policy. For the past several decades, this understanding has been greatly shaped by the New Keynesian literature, leading to what Clarida et al. (1999) call the “science of monetary policy.” In this framework, the policy interest rate set by the central bank and its commitment to vary that rate consistently with its objectives play a critical role in determining the effects of monetary policy.

The precise channel through which monetary policy influences the economy has been a debatable issue. In typical transmission models, given assumptions of frictionless financial markets, perfect asset substitutability and rational expectations, the overnight rate set by the central bank determines the long-term interest rate, the exchange rate, and other asset prices which, in turn, determine the path of aggregate spending and inflation (Taylor 1995; Woodford 2003; Boivin et al. 2010). Term and risk premia—central to the analysis of imperfect asset substitutability by Tobin (1969) and Modigliani and Sutch (1967)—therefore play no role in the transmission mechanism. In addition, these models assume that size of the central bank balance sheets has no independent influence on aggregate demand so that bank reserves are provided perfectly elastically at the policy rate. Another assumption underlying these monetary transmission models is that in globally integrated economies, a central bank’s ability to control interest rate is a function of the degree of exchange rate flexibility (the so-called trilemma doctrine ). Hence, a fully floating exchange rate is able to insulate domestic monetary policy from external shocks (see Clarida et al. 2001; Gali and Monacelli 2005; Woodford 2009).

The New Keynesian models, particularly those incorporating features such as asymmetric information and credit market imperfections, appeared to describe fairly well the working of monetary policy before the 2008 financial crisis. For instance, the “financial accelerator” literature (e.g., Bernanke et al. 1999; Bernanke and Gertler 1995) highlighted the role of “external finance premium” in the transmission mechanism. A key point stressed by these papers was that the external finance premium paid by the borrowers varies with the interest rate, depending on the quality of their balance sheet. Because collateral is central to households and firms’ ability to access credit, asset prices play a major role in the amplification of monetary shocks (see Kiyotaki and Moore 1997). In contrast, the “credit view” literature stressed the importance of lenders’ capital and financing constraints which affect their ability to supply credit (Kashyap and Stein 2000; Bean et al. 2002). Given imperfect substitutability between reservable deposits and other liabilities, monetary policy generates credit supply effects because some banks are less able than others to replace deposit funding with outside finance.

However, the developments since the 2008 crisis have led to a major reassessment of the mainstream monetary transmission models. While integrating real-world financial frictions into monetary policy models continues to be a tough challenge for economists, recent research has pointed to, at least, four directions where the changes have been very significant. First, the crisis has demonstrated that central banks can use both quantity and price instruments simultaneously to achieve their goals. This has been most visible in the use of balance sheet policies by major advanced economy central banks to control monetary conditions after the short-term interest rate hit the zero lower bound. As Friedman and Kuttner (2011) note “the ability (of the central bank) to choose the level of the policy interest rate and the size of its balance sheets independently, over time horizons long enough to matter for macroeconomic purposes…represents a fundamental departure from decades of thinking about the scope of central bank action.” In contrast to the efficient market models underlying the term structure hypothesis, there is now explicit recognition that term and risk premia play a crucial role in the determination of the cost of credit even in financially mature economies. The recent analysis by Gertler and Karadi (2013) has reinforced this view.

Second, in contrast to what the conventional monetary transmission models assumes, there is now increasing evidence that long-term interest rates are influenced more by global factors than local factors such as domestic business cycle or monetary policy (Obstfeld 2015; Turner 2014, 2015; Miyajima et al. 2015). While the tendency of the long-term interest rate to move together across economies is nothing new, what is special is that the correlation of bond yields has increased significantly after the 2008 financial crisis for emerging markets. Such a shift in bond market correlation assumes importance because it can reduce the role of the policy rate in the transmission mechanism and contribute to unwarranted fluctuations in credit, creating risks to monetary and financial stability.

Third, there has been a clear shift in the perception about the role of the exchange rate in the transmission mechanism. Not only have the responses of trade variables to exchange rate changes been smaller than assumed earlier, but exchange rates have also become far more volatile than can be predicated by measures of interest rate differentials. A further dimension has been that the growth of currency mismatches associated with the expansion of unhedged dollar borrowing by EMEs means that currency depreciation can be contractionary (Bruno and Shin 2014).

Finally, a key missing link in the earlier literature, as recently documented by Gertler and Kiyotaki (2011), was that it largely focused on the financing constraints on the nonfinancial borrowers and treated the financial intermediaries as a veil, thus ignoring the numerous agency problems and nonlinear asset price dynamics confronting the financial system. Indeed, as shown by Adrian and Shin (2010a, 2010b), capital and value-at-risk constraints facing financial intermediaries matter for their lending behavior. Because monetary policy affects asset prices and bank profitability, it can alter these constraints, causing shifts in credit supply. The interaction between the short-term interest rate, lenders’ risk perception, and their attitude toward lending has been increasingly referred to as the “risk–taking channel” of monetary policy (Borio and Zhu 2012; Bakaert et al. 2013).

In what follows, in Sect. 2 we first start with a brief review of financial intermediation in EMEs to highlight the fact that many of the recent developments in the monetary transmission mechanisms can be traced to changes in the size and the nature of financing in EMEs as well as the external monetary environment facing them.

In Sect. 3, we discuss a few implications of these changes for the role of the interest rate, exchange rate, and credit channels in EMEs. One key finding of this section is that domestic monetary policy has to contend with increased globalization of debt markets and long-lasting shifts in global long-term interest rates and investor risk appetite that have the potential to make monetary conditions very volatile.

In Sect. 4, using a structural VAR model, we consider some empirical applications to India and note that the relative closed character of India’s domestic bond markets has probably helped to limit the impact of external monetary shocks on the economy, particularly through the bond price and exchange rate channel. However, both equity price and credit channels continue to remain quite active.

In Sect. 5, we turn to a reduced from monetary transmission model to demonstrate a few policy challenges for the central bank when domestic bond markets are closely linked to the international bond markets. A key implication is that, with globalization of debt markets, the conduct of monetary policy through short-term interest rate has become a much more complicated affair, raising issues about the appropriate instruments for stabilizing inflation and output. In Sect. 6 we present conclusions.

2 Recent Changes in Financial Intermediation in EMEs

Historically, banks have been at the center of financial intermediation in EMEs. In addition, in the earlier decades, even though the financial systems of EMEs were open to international portfolio flows, the scale of these flows remained relatively limited in many cases, which meant that domestic interest rates were, to a large extent, tightly linked to the key monetary policy instruments of the central bank. Hence, monetary policy effects were largely determined by developments in the banking system. However, the environment in which monetary policy is conducted in EMEs has undergone major changes over the past decade. In this section we focus on three such changes: (a) the relative role of banks versus debt markets, (b) globalization of debt markets, and (c) the evolution of global long-term interest rates.

2.1 The Relative Role of Banks and Bond Markets

Table 1 shows total credit extended to the nonfinancial private sector in major emerging Asian economies as a percentage of GDP before and after the 2008 financial crisis as well as in mid-2000s. The data covers credit from all sources, including banks and bond markets from domestic and foreign sources. For comparison, the table also provides averages for other regions. As can be seen from the table, the ratio of total credit to GDP has increased rapidly in most countries between 2004 and 2013. The trend started earlier but accelerated following the 2008 financial crisis.

Table 1 Private sector credit and domestic bank lending in EMEs^a

It is important to note that credit has grown much faster in economies that are more open to capital flows and/or maintain some form of exchange rate link with currencies of major advanced economies than those that are less so financially open or have adopted a flexible exchange rate regime. This is particularly true in Hong Kong SAR, with its linked exchange rate system and highly open capital account as well as its role as an international financial centre, but also China even with its relatively closed capital markets. Notwithstanding their relatively independent monetary policy regimes, Korea, Malaysia, and Singapore have all seen rapid increases in their total credit to GDP ratios since the crisis.

Another fact emerging from Table 1 is that the share of credit from the banking system in total nonfinancial private credit has fallen in a number of countries. Even though banks continue to be important in credit allocation in EMEs, their role has declined over the past decade, especially in Asia. China is a major example where the share of bank credit in total credit has fallen by 21 percentage points between 2004 and 2013. Many other Asian economies have also seen significant declines in the share of bank credit .

A mirror image of the declining share of banks in credit is the growing importance of the bond market. Graph 1 shows two main dimensions of debt securities issuance by EME nonfinancial corporations—domestic and international issuance. There is evidence that financial intermediation through bond markets has increased, and a large part of that intermediation has moved offshore.¹ What is striking is that EME nonfinancial corporations have sharply increased their international debt issuance, which registered more than threefold growth between 2008 and 2013. Again, Asia seems to be leading the EMEs.

Graph. 1

Domestic and international debt securities. Amounts outstanding, in trillions of USD

2.2 Globalization of Debt Markets

In addition to changes in the financing structure, the markets for debt securities have become increasingly global. There are several dimensions to the recent globalization of debt markets . They relate to the diminishing importance of national borders in the determination of capital flows, the use of currency in the denomination of debt transactions, and the structure of EME local currency debt markets.

In the traditional definition of capital flows, reported by the IMF, the concept of residency of the borrower plays a central role in the determination of economic area of a country and hence the magnitude of flows into and out of that country. However, as pointed out by Bruno and Shin (2014); Avdjiev et al. (2015), with capital flows straddling national borders, residency as a concept for measuring capital flows has become increasingly irrelevant. Take, for instance, a subsidiary of a Brazilian firm located in London issuing a dollar bond in London. This will not be reckoned as capital flow in the balance of payment statistics even though the funds may be ultimately used by the parent firm in Brazil. Avdjiev et al. (2014) discuss several channels through which the funds mobilized by the subsidiaries could appear as disguised capital flows. These funds could either be lent directly to the parent company as within-company loan or be extended as credit to another company in the same country or be simply parked as cross-border deposit in the domestic banking system .

The red line in left-hand panel of Graph 2 shows the scale of outstanding debt issuance by EMEs by nationality of borrowers. These numbers therefore capture international debt issuance by all nonfinancial corporations of a country residing anywhere in the world and are thus different from those based on the residency in Graph 1 (or shown by the blue line in Graph 2). On this definition, debt issuance by EMEs nonfinancial firms has not only grown rapidly since 2009, but they are now twice as large as those based on the residency of borrowers .

Graph. 2

Global debt markets and US dollar credit

The second dimension of globalization of debt markets concerns the use of national currencies. An implicit assumption in the traditional monetary transmission models is that national balance sheets are denominated in national currency so that changes in monetary policy have implications for the flow of funds within the economy.² However, as the experience of widespread dollarization in the 1980s and 1990s demonstrated, the influence of national monetary policy is limited when a large part of domestic liabilities and assets are denominated in foreign currency (e.g., Kamin et al. 1998; Mohanty and Turner 2008).

While the degree of dollarization of the EME banking system has fallen considerably over the past decade that of the nonbank sector has increased. This is a global phenomenon but with a large EME component. McCauley et al. (2015) estimate the outstanding dollar debt of nonbank borrowers outside the United States. In other words, these are not dollar borrowing by US residents that are naturally affected by the dollar interest rate but by nonbank borrowers in the rest of world that have chosen to denominate their debt in dollar. As can be seen from the right-hand panel of Graph 2, total dollar credit outstanding against the nonbank, non-US borrowers expanded by more than fourfold between 2000 and 2015, from less than $2.2 trillion to $9.7 trillion. Dollar credit to EME non-bank borrowers has recorded the fastest increase, constituting the single largest component of total by 2015 .

Reinforcing this trend is the third dimension of globalization of debt markets linked to internationalization of EME bond markets . During the 1980s and the 1990s, the EME local currency bond markets were not only underdeveloped but remained largely inaccessible to foreign investors. This, however, started to change in the beginning of the 2000s, as local bond markets started to develop in many EMEs and foreign investors preferred to invest in these markets, reducing barriers to international arbitrage. Estimates by the World Bank suggest that the share of nonresident holding of EME local currency bonds in total stock has more than doubled between 2008 and 2013 (from 13 % to 30 %). According to a BIS survey conducted in 2012, in a number of major EMEs these shares varied from 30 to 50 % (Mohanty 2014). Indeed, as argued by Shin and Turner (2015), growing nonresident investment in EME local currency bonds and the rapid expansion of international debt issuance by EME corporations represent two defining elements of the new financial landscape in EMEs .

2.3 Global Long-Term Interest Rates

Finally, another major factor shaping monetary conditions across the world has been the behavior of the global long-term interest rate. The left panel of Graph 3 plots King and Low’s (2014) estimate of global real long-term interest rate which is an average of real 10-year spot yields of G7 economies (nominal yield minus expected inflation). The red and blue lines show the unweighted and GDP-weighted averages, respectively. Whereas the world real long-term interest rate was range bound during 1980s and early 1990s, it started to decline steadily in the beginning of the 2000s. The trend accelerated after the 2008 financial crisis, particularly following the introduction of large-scale asset purchase programs by the Federal Reserve and other advanced economy central banks . GDP-weighted real long-term interest rate tells a similar story, although data are available for a relatively short period .

Graph. 3

World real long-term interest rates. In per cent

Evidence reported in a recent report by the Executive Office of the President (2015) suggests that the US long-term yields tend to revert to a mean over time, but that reversion can be slow and not necessarily to a constant mean (Hamilton et al. 2015). Economic theory suggests that real interest rates are likely to be bounded because the underlying variables such as saving and investment respond to changes in interest rates—so bringing them back to their steady-state levels. However, the fact that the world real long-term interest rate has been declining over much of the past two decades confirms the hypothesis that the changes in either direction can be quite persistent. This can cause major shifts in resource allocation, international capital flows, and spending across countries.

To understand the sources of this variation, the right-hand panel of Graph 3 shows an estimate of US 10-year term premium taken from Hordahl and Tristani (2014). There are, of course, other components of the long-term interest rate, viz., market expectations of the short-term interest rate and inflation expectations, which are not reported here. The graph, nevertheless, shows that a significant part of the recent decline in the long-term interest rate reflected movements in the US term premium, which , after trending down during much of the past two decades, has fallen to very low or negative levels.³

3 Monetary Transmission Mechanisms Post-2008 Crisis

How have these changes affected transmission of monetary policy in EMEs? In this section, we consider three main channels—the interest rate, the exchange rate, and the credit channel—to review the potential effects of recent changes in financial intermediation on the transmission mechanism.

3.1 The Interest Rate Channel

Interest rate often plays a key role in the transmission of monetary policy shocks. A rise in the policy rate by the central bank to dampen incipient inflation pressure leads to a rise in the short-term market interest rate and therefore most borrowing and lending rates in the economy. For this, the real interest rate is important: a rise in the nominal rate that reflects higher inflation expectations—so that real rate remains constant—will not change the perceived marginal costs of borrowing. Furthermore, since monetary policy operates most effectively by influencing the demand for durable goods, what matters is the extent to which changes in the policy rate affect funding costs for long-term projects. Following Mishkin (2007) and Boivin et al. (2010), this relationship can be formalized by a user cost of capital equation, which, in a closed economy, can be expressed as follows:

$$ {\rm{U}}_{{\rm{t}}}^{{\rm{c}}} = {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} [{\rm{E}} \{ \left( {{\rm{i}}_{{\rm{t}}}^{{\rm{m}}} - {{\uppi}}_{{\rm{t}}} } \right) - ({{\uppi}}_{{\rm{t}}}^{{\rm{c}}} - {{\uppi}}_{{\rm{t}}} )\} + {{\updelta}}] $$

which can be equivalently written as

$$ {\rm{U}}_{{\rm{t}}}^{{\rm{c}}} = {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} [{\rm{E}} \{ {\rm{i}}_{{\rm{t}}}^{{\rm{m}}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} \} + {{\updelta}}] $$

(1)

where ${\rm{P}}_{{\rm{t}}}^{{\rm{c}}}$ is the relative price of new capital, ${\rm{i}}_{{\rm{t}}}^{{\rm{m}}}$ is the domestic short-term interest rate, ${{\uppi}}_{{\rm{t}}}$ is the inflation rate, ${{\uppi}}_{{\rm{t}}}^{{\rm{c}}}$ is asset price inflation, δ is the rate of depreciation, and E is the expectation operator. We are abstracting away from tax considerations, which nevertheless may be important sometimes, for example, when thinking of deductibility of the interest rate by adjusting the nominal interest rate by the marginal tax rate. The user cost equation shows that spending decisions of the agents depend on the expected real interest rate and real price appreciation of the asset over its entire life time. Assuming sticky prices, monetary policy affects demand for long-lived assets to the extent that it can change the expected future path of the real interest rate and the value of the asset. It is therefore obvious that the long-term interest rate plays a key role in the transmission mechanism of monetary policy .

Housing investment is a clear example of how the user cost channel works. A tighter monetary policy increases the cost of capital for the prospective home buyers both by increasing the long-term financing costs and weakening the expected future house price appreciation, causing a slowdown in the construction activity and aggregate demand. This direct effect is magnified by the fact that developments in the housing market affect the wealth position and creditworthiness of borrowers. For instance, in the United States, residential investment is found to be highly sensitive to the user cost of capital, even though the estimates of elasticity vary widely , from −0.2 to −1.0 (Mishkin 2007).⁴

The above user cost framework assumes that long-term interest rates and asset prices move in tandem with the expected future path of the short-term interest rate. However, to the extent that the term premium may move independently—as the events following the 2008 global financial crisis demonstrated—long-term funding costs can deviate substantially from the stance of monetary policy. In addition, Eq. 1 was written in the context of a closed economy, but, as the previous section highlighted, this assumption is increasingly unrealistic with the growing global integration of EMEs’ debt markets .

One way to account for these factors is to bring them explicitly into the user cost equation. Let us denote the long-term sovereign bond yield in the domestic and international markets as ${\rm{LT}}^{{\rm{d}}}$ and ${\rm{LT}}^{{{\rm{US}}}}$, respectively, and noting that international arbitrage implies that the expected rate of depreciation of exchange rate should be equal to the sum of yield differentials and currency risk premium ${\varvec{\uprho}}$, we have:

$$ {\rm{LT}}^{{\rm{d}}} - {\rm{LT}}^{{{\rm{us}}}} = {\rm{E}}\left[ {\Delta {\rm{e}}} \right] + {\varvec{\uprho}} $$

(2)

Decomposing the long-term yields into expected interest rate and term premium such that ${\rm{LT}}^{{\rm{d}}} = {\rm{E}} \left[ {{\rm{i}}_{{\rm{t}}}^{{\rm{m}}} } \right] + {\rm{q}}^{{\rm{d}}}$ and ${\rm{LT}}^{{{\rm{US}}}} = {\rm{E}} [{\rm{i}}_{{\rm{t}}}^{{{\rm{us}}}} ] + {\rm{q}}^{{{\rm{us}}}}$ and substituting in Eq. (1), we have:

$$ {\rm{U}}_{{\rm{t}}}^{{\rm{c}}} = {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} [{\rm{E}} \{ {\rm{i}}_{{\rm{t}}}^{{{\rm{us}}}} +\Delta {\rm{e}} - {\varvec{\uppi}}_{{\rm{t}}}^{{\rm{c}}} \} + ( {\rm{q}}^{{{\rm{us}}}} - {\rm{q}}^{{\rm{d}}} ) + {\varvec{\uprho}} + {\varvec{\updelta}}] $$

(3)

According to this equation, the user cost of capital in an open economy depends on three main elements: the first is the degree of correlation between the risk-free domestic and international yields, which could stem from the correlation of the excepted future EM short-term interest rates with the expected future fed funds rate (assuming that the US interest rate is the base rate for all EMEs). In a world with prefect capital mobility, the domestic risk-free long-term interest rate equals the US risk-free long-term interest rate, and monetary policy primarily works through the exchange rate. The second element is the degree of correlation between EME term and currency risk premia with the US term premium. Again, under perfect capital mobility, it is the US term premium that matters for the EME long-term interest rate , plus a country bond risk premium and a currency risk premium . The third element is the asset price change associated with capital flows, which also affects the user cost of capital. Note that Eq. (3) is an expression linking the cost of credit with the interest rate from the perspective of the borrower, and hence does not reflect the factors that may affect the supply of credit .

3.1.1 Correlation of Bond Yields

A key empirical issue is how domestic funding costs actually respond to a change in central bank’s policy rate. To the extent that EME firms have unrestricted access to international debt markets, the pass-through of the policy rate to domestic borrowing costs could be reduced because the user cost of capital is likely to move closely with foreign interest rates. In this case the impact of domestic monetary policy depends on the degree of substitutability between domestic currency assets and dollar assets. Assuming limited exchange rate changes, a policy-induced rise in interest rate would prompt borrowers to switch to dollar debt and savers to domestic currency assets, leading to a partial loss of control of monetary conditions. At the same time, monetary authorities must consider the adverse implications of interest rate changes for the exchange rate and financial stability (see Rossini and Vegas 2008).

Ideally, the user cost of capital should not change if firms borrowing in dollars hedge their exposure to the expected future depreciation of the domestic currency against the dollar. In practice, however, such hedging is unlikely to be complete, and firms may be attracted to minimize funding costs in the short run by leaving a large part of their dollar borrowing unhedged.

In partially dollarized economies, much depends on how domestic yield curves behave in response to domestic and foreign interest rate shocks. As a first pass, in the left-hand panel of Graph 4 we report the coefficient of rolling correlation of EME interest rates with the US interest rate and US term premium. The correlations are computed using monthly data over a fixed window of 3 years for a group of major EMEs but excluding those that have a fixed exchange rate regime (e.g., Hong Kong SAR). The solid red line shows that the correlation of average EME policy rate with the fed funds rate has fluctuated over time, with a mean close to zero for the period shown. In the postcrisis period, this correlation has been actually negative. Because the fed funds rate has been close to zero since 2009, we recomputed the correlation using an estimate of the shadow fed funds rate, taken from Lombardi and Zhu (2014). As the dotted red line shows, the results are broadly similar, although the correlation has recently turned positive.

Graph. 4

Correlation of interest rates

In contrast, as the blue line shows, with a few exceptions, the correlation of EME long-term interest rates with US long-term rate has not only been positive in the past decade, but it also increased steadily following the 2008 crisis. What is striking is that this correlation appears to stem mostly from the comovement of EME long-term rates with the US term premium (yellow line). The right-hand panel of Graph 4 throws further light on this by reporting the first principal component of EME and advanced economy long-term interest rates, this time including a larger pool of countries than just the United States. The common component of the two series moved in the opposite direction to each other before the crisis but started to comove very tightly after the crisis. It also broadly confirms King and Low (2014) estimates that the global long-term rate has declined to very low levels in the past decade .

Since interest rate levels are likely to be correlated because of several factors unrelated to monetary policy, a formal test must consider these correlations at first differences and allow for other determinants. A familiar test proposed by Sambaugh (2004); Klein and Sambaugh (2013) is as follows:

$$ \Delta \varvec{i}_{{\varvec{jt}}} = \propto + {\beta}\Delta \varvec{i}_{{{bt}}} +{\gamma}^{{\prime }} \varvec{x}_{{{jt}}} + {u}_{{{jt}}} $$

(4)

where i is either a short-term or long-term interest rate, the subscript “j” refers to home country and “b” to a base country, x is a vector of domestic variables determining the home interest rates, and u represents the difference in risk characteristic of home and base country assets. In a fully credible peg regime, the home country interest rate equals the base country interest rate; hence, ${\beta}= 1$ and ${\gamma}= 0$. Conversely, a fully independent monetary policy implies that ${\beta}= 0$ and ${\gamma}= 1$. For any intermediate values of $\beta$ and $\gamma$ the pass-through of the base country interest rate to home country rate will be partial .

Recent studies investigating Eq. (4) have generally converged to the conclusion that $\beta$ is significantly positive for long-term interest rates but insignificant or only weakly positive for the short-term interest rates⁵. Miyajima et al. (2015), using data for a panel of 11 well-developed EME local currency bond markets, found that the response of EME 10-year bond yields to US 10 Treasury yields increased sharply to 53 basis points (due to 100 basis points increase in US treasury yield) after the 2008 crisis from 31 basis points for the entire sample starting January 2000. During periods of adverse market dynamics (such as the May–June 2013 “taper tantrum”), this response rises to slightly over 100 basis points. Using quarterly data for the most recent periods and a larger set of EMEs, Sobrun and Turner (2015) report similar results: whereas EME bond yields were weakly correlated with US yields during 2000–2004 that response became strong and statistically significant after 2005.

A litmus test for many studies is how to control for the unobserved common shocks that could lead to spurious correlations of interest rates. Obstfeld (2015) addresses this issue by considering different base country rates for different countries (such as the dollar interest rate for Mexico, the euro interest rate for Poland, and so on) so as to minimize the common time effects in the panel regression. His results suggest that while the coefficient of the short interest rate in Eq. (4) is small and insignificant, that of the long-term yields is highly significant at 1 % level. Even after changing the base country rates, the response of EME long-term rates to advanced economy long-term rates continues to be 40–50 basis points .

Kharroubi and Zampolli (2015) use a cross-section mean group estimator, as suggested by Pesaran (2006), to control for unobserved common shocks. Their results suggest that short-term interest rates in the flexible exchange rate regimes neither respond to the base country interest rate nor to global risk cycles. However, they find statistically significant effect for the domestic long-term interest rates, which rises by 60 basis points in response to 100 basis points rise in the base country long-term rates. In addition, their estimates suggest that the response of domestic long-term interest rates to domestic short-term rates is relatively small (around 20 basis points in both the pre- and post-crisis period).

In sum the evidence is quite solid that long-term interest rates of EMEs have been highly correlated with global long-term rates, consistent with our open economy user cost of capital framework. In addition, some studies have shown that this correlation could be due more to the US term premium than market expectations of the US short-term interest rate (Miyajima et al. 2014). A shock to the US term premium is qualitatively different because it has the potential to generate more severe repricing of EME assets.⁶

3.1.2 Pass-Through to Bank Lending Rates

Is the interest rate channel still relevant? The answer depends, of course, on the structure of the financial system of a country. While the ratio of bond financing in total credit has increased across EMEs, there is significant difference across countries. Yet, a high degree of bank financing does not necessarily insulate domestic monetary policy from external shocks because banks are both issuers and investors in the bond market and compete with the bond market for their clients.

The degree of response of bank lending rates to the central bank policy rate will be conditioned by several factors. The first is the degree of competition within the banking system as well as with the bond market. In general, the higher is the degree of competition among banks in the loan market, the lower the probability that the bank intermediation spreads would fluctuate to offset the impact of policy rate change.⁷ Introduction of more players in the credit market through bond markets can reduce the oligopolistic structure of the banking system, leading to stronger transmission of monetary policy to the banking system. On the other hand, the greater importance of capital markets in financial intermediation may accentuate information asymmetry problems between borrowers and lenders, leading to higher risk premia and a weaker monetary transmission mechanism , more generally.

A second factor is the funding structure of the banking system . Bank lending rates reflect expected short-term rates over the full maturity of loan, so include a maturity risk premium that can vary with the health of banks’ balance sheets. In addition, banks have a more varied liability structure than just reservable deposits which suggests that their average funding costs may change only slowly in response to a change in the central bank policy rate.⁸ This factor assumes particular importance, given that banks in many EMEs have accessed non-deposit funding sources, including the bond market, to finance a significant part of their asset growth. A recent BIS survey indicated that average contribution of non-deposit liabilities in total bank liabilities in a group of 20 EMEs was about 28 % during 2004–2013 (Ehlers and Villar 2015). In countries where financial markets are more developed (e.g., Hong Kong SAR, Korea, and Mexico) such non-core liabilities accounted for a much larger share of total bank liabilities .

A third factor is the nature of deposit and loan contracts . A high share of short-term liabilities in the total bank liabilities and short-term loans in total bank assets increases the pass-through of a given change in the policy rate to average funding costs of banks and, ultimately, to those of their borrowers. That said, the reliance on short-term liabilities also makes banks more vulnerable to shocks to money markets and capital flows, reducing their ability to transform maturity and sustain credit supply. This implies that banks in countries with an underdeveloped bond market are likely to face a trade-off while optimizing their asset and liability structure to minimize funding and interest rate risks.

This factor is likely to be particularly important in EMEs where contractual maturity of bank liabilities tends to be quite short, with a median of just about 4 months for a group of 11 economies at the end of 2013 (Ehlers and Villar 2015). In addition, a high proportion of bank deposits in EMEs bears variable rates contracts (50–70 %), which means that deposit rates in many cases are effectively indexed to the policy rate. Although information is more limited for bank lending contracts, the picture is somewhat different across regions. For instance, while variable residential mortgage contracts accounted for 70–99 % of total residential mortgages in Asia in 2013, fixed rate contracts dominated mortgage markets in Latin America, with a ratio close to 100 % in Brazil and 70–96 % in Argentina, Chile, Colombia, and Mexico.

A rough indication of how bank lending rates behave in response to policy rate changes is given by the scatter plots in Graph 5 summarizing data for seven Asian economies over the past decade. The preference for four-quarter change to single-quarter change in interest rate was guided by the consideration that bank lending rates exhibit short-run stickiness due to the existence of fixed adjustment costs. The positive slope of the trend line in Graph 5 suggests that the response of the lending rate to monetary policy is quite different from that of the long-term bond yield. That being said, the average response of the household lending rate (16 basis points) is just about half of the response of the business lending rate (34 basis points). In addition, the explanatory power of regression is not very high for household lending rates .

Graph. 5

Lending rates and policy rates. Annual changes, in percentage points

A similar exercise exploring the relationship between bank lending rate and the US long-term rates did not yield meaningful results.⁹ While more systematic analysis is needed to reach reasonable conclusions, these preliminary evidences nevertheless suggest that the interest rate channel of monetary policy may not have been completely eroded by the recent rapid growth of dollar borrowing by EME firms .

3.2 The Exchange Rate Channel

Another important transmission channel is the exchange rate which mainly operates in economies with a flexible exchange rate. As interest rate falls due to expansionary monetary policy, domestic interest-bearing assets become relatively less attractive, triggering capital outflows, and exchange rate depreciation. However, currency depreciation may have several opposing impacts and the net effect on output may turn out to be either positive or negative. While depreciation may boost exports and hence overall aggregate demand, on the one hand, it may also mean erosion in net worth for the borrowers with foreign currency debt and thus a decline in aggregate spending, on the other. A depreciating currency can also lead to higher inflation depending on the degree of pass-through of import prices into domestic prices.

A good example of how the exchange rate channel works is Singapore —an open economy par excellence. Given a high import content of domestic consumption (around 40 %), the exchange rate has a direct impact on domestic inflation (Loh 2014). And, since exchange rate has predictable effects on the demand for exports and factor inputs, it also has an indirect effect on inflation. In addition, the country has a large net international investment position vis-à-vis the rest of the world, and the daily exchange rate movement of the Singapore dollar is managed by the Monetary Authority of Singapore. With trade effects reinforcing balance sheet effects—currency depreciation improving rather than worsening net wealth position—the exchange rate plays an important counter-cyclical role in the economy .

However, in economies with significant currency mismatches, the role of the exchange rate can be very different. For instance, consider the following aggregate demand equation:

$$ {y}_{{t}} - {y}^{{*}} = {\gamma }\left( {{y}_{{{t} - 1}} - {y}^{{*}} } \right) - {\beta }\left( {{r}_{{t}} - {r}^{{*}} } \right) - { \lambda }\Delta {e}_{{t}} + \epsilon_{{t}} $$

(5)

where ${y}^{}$ is the actual output, $\varvec{y}^{\varvec{*}}$ is the potential output, r is the real interest rate, $\varvec{r}^{\varvec{*}}$ is a normal or equilibrium real interest rate, e is the real exchange rate, and $\epsilon$ is a disturbance term. Output gap in this model is negatively related to the real interest rate gap as well as the exchange rate. A negative coefficient on $\varvec{ \lambda }$ assumes that currency depreciation is associated with improved trade balances and easier financing conditions. In practice, however, $\varvec{ \lambda }$ could take any plausible value depending on the structure of the economy. One form of contractionary devaluation, highlighted by the early literature, is exchange rate-induced rise in import costs, which turns $\varvec{ \lambda }$ positive, particularly in economies heavily dependent on commodity imports (Frankel 2011). It is also possible that exchange rate elasticities are considerably small in economies that rely heavily on imported inputs for export production. The second type of contractionary devaluation that received much attention during the 1990s EME currency crises is the case of liability dollarization where currency depreciation is associated with widespread deterioration of balance sheets of borrowers, causing tighter financing conditions.¹⁰

Eichengreen (2002) points out that the impact of the exchange rate in economies with large currency mismatches is likely to be nonlinear—while small currency depreciations are likely to satisfy the conditions for $\varvec{ \lambda }$ being negative, large depreciations can cause severe financial distress because “they confront banks and firms with asset prices for which they are unprepared, while doing little to enhance competitiveness effects because of the speed with which they are passed through into inflation.”

However, scepticisms about the role of the exchange rate changed considerably in the aftermath of the 1994–1995 Mexican crisis and the 1997–1998 Asian financial crises , which not only heralded a new era of independent monetary policy in EMEs—led in many cases by the introduction of inflation targeting—but led to concerted efforts by the EM authorities to reduce the degree of currency mismatches (see BIS 2008). At the same time, a significant reduction in the degree of exchange rate pass-through into inflation meant that the exchange rate improved the growth and inflation trade-off facing the central bank.

3.2.1 Post-Crisis Changes

Since 2010, however currency mismatches in many EMEs have increased notably because of a substantial increase in foreign currency borrowing by emerging market non-financial companies (Chui et al. 2016). The exchange rate therefore has come back to the center of monetary policy debate post-2008 crisis. As shown in Graph 6, real exchange rates have exhibited protracted cycles with upswings of currency appreciation followed by downswings of depreciation. Latin America is a case in point. While average real exchange rate in the region at end-October 2015 has been roughly at its level in 2005, the intervening period has seen rapid movements on both the strong and weak sides. In Brazil, exchange rate cycles have been associated with large changes in the financial and economic conditions. A similar, though less protracted, trend in the real exchange rate has been visible in several parts of Asia (e.g., India and Indonesia) and central and eastern Europe (e.g., Turkey and Hungary).

Graph. 6

Real exchange rate in emerging markets¹

A key contributing factor has been the behavior of commodity prices (BIS 2014). In many oil-exporting countries, exchange rate swings have been associated with protracted shifts in the terms of trade. At the same time, the correlation between EM exchange rates with indicators of global risk aversion (such as the VIX) increased considerably in the aftermath of the 2008 global crisis (see, Rajan 2014); Miranda-Agrippino and Rey (2013) and Rey (2013) have argued that shifts in global investors’ risk appetite have led to an unusual convergence of exchange rate and asset price cycles across the globe.

One effect of such currency movements in the face of large accumulated foreign currency debt has been that the EME credit cycles now tend to comove more closely with the dollar exchange rate. Bruno and Shin (2014) have used the expression “the risk taking channel of the currency appreciation.” In Bruno and Shin model, global banks play a key role in the transmission of external shocks because they channel liquidity from the US money market to the local EME banking system. Since dollar depreciation improves balance sheets of bank borrowers with dollar debt, it reduces the effective credit risk facing banks, leading to expansion of lending. Conversely, the periods of large dollar appreciations are followed by contractions in global banks’ balance sheets and widespread dollar shortages. In terms of Eq. (5), this means that the value of $\varvec{ \lambda }$ is just not a static function of the degree of currency mismatches but varies depending on the force of the amplification mechanism at work .

There is also increasing evidence that the exchange rate can affect financing conditions even without such financial imbalances. One such mechanism is the amplification of market volatility stemming from growing interaction between the foreign exchange market and the bond market that can, at times, be triggered by speculative investor positioning and pro-cyclical investment strategy pursued by some investors such as professional asset managers (Feroli et al. 2014). Turner (2012) shows that the hedged and unhedged returns on EM local currency bonds have consistently diverged, suggesting that the exchange rate played a crucial role in the bond market dynamics. The failure of the uncovered interest parity means that foreign investors have affected the risk premium, causing large fluctuations in domestic monetary conditions.¹¹ When the exchange rate appreciates investors may take speculative carry positions in bond markets to gain from the expected future appreciation, which drives down EM risk premium and bond yields to very low levels. During the periods of market stress, however, as currency depreciates and uncertainty about the future exchange rate rises, foreign investors rush to exit, causing higher bond yields and tighter financing conditions.¹²

Such currency carry trades may not be restricted only to nonresident investors. EME residents could also make use of dollar debt issuance to undertake similar investment strategies, leading to volatile capital and credit flows. For instance, a recent study has found that nonfinancial companies had used US dollar bond issuance to take on financial exposure that shared attributes of dollar carry trades (Bruno and Shin 2015). The proceeds of such bond issuance were invested in high-yielding bank deposits as well as in shadow banking products and commercial papers.

3.3 The Credit Channel

The credit channel of monetary policy operates through the balance sheets of the lenders and the borrowers, depending on the degree of financial imperfections in an economy.¹³ A key question raised by the 2008 crisis, in the light of the recent rapid increase in bank leverage in many advanced economies, is the extent to which the behavior of financial intermediaries can contribute to magnifying monetary policy effects on credit supply. Gertler and Kiyotaki (2011) introduce financing constraints facing banks into the original financial accelerator models proposed by Bernanke et al. (1999). To the extent that banks’ own balance sheet conditions constrain their ability to access deposit funding and they are vulnerable to idiosyncratic liquidity shocks, dysfunction in either market leads to jumps in the external finance premium, as experienced by many countries during the 2008 financial crisis .

An important aspect of credit channel that has received much attention post-2008 crisis is the link between monetary policy and risk taking by financial intermediaries . Adrian and Shin (2010b) consider the behavior of financial intermediaries that typically mark a significant part of their assets to market (shadow banks such as interdealed broker or banks that invest heavily on securitized products). In their model, individual bank managers face value at a risk limit and are risk neutral. An easier monetary policy, boosting asset prices and profitability, reduces banks’ capital, and value-at-risk constraints, encouraging them to take on more risk. In the equilibrium, the market price of risk becomes endogenous, amplifying the impact of monetary policy on credit.

Acharya and Naqvi (2012) discuss a similar risk taking channel where the incentive structure facing bank mangers rather than financing constraints of banks plays a key role in the propagation of credit and asset price cycles. Since bank managers’ compensations are linked to the volume of loans, excessive liquidity is associated with systematic mispricing of downside risk, causing credit, and asset price bubbles. Nicolo et al. (2010) discuss several other mechanisms that may be at work in strengthening the link between monetary policy and risk taking by financial intermediaries.

Given a high degree of imperfection in their financial systems, the credit channel may be of particular relevance to the EMEs. Several recent studies based on the pre-crisis EME data suggest that monetary policy has a stronger effect on the banking systems that are less well-capitalized and competitive than others,¹⁴ and monetary policy may have particularly large effect on smaller banks whose access to outside finance may be very limited. Agenor and Montiel (2008) discuss monetary policy effects in economies where weak lenders’ protection right and high default probability cause banks to over-collateralise loans. Under such conditions, an easy monetary policy lowers lending constraints of large firms but squeezes small and marginal borrowers who have little to gain from higher asset values. A segmented credit market with a large informal sector thus makes EMEs simultaneously vulnerable to pro-cyclical credit market dynamics and active credit rationing. Agenor and Pereira de Silva (2013) cite the example of Brazil where bank lending spreads have been inversely correlated with fluctuations in economic activity, providing evidence on the pro-cyclical credit dynamics .

While it is difficult to determine the precise effects of the changes in financial intermediation on credit supply in EMEs , recent research does seem to provide useful guides. One key finding emerging from this literature is that globalization of banking may weaken the link between domestic monetary policy and credit variables (Cettorelli and Goldberg 2012). Because global banks use their own internal capital market to channel funds across borders, they could potentially offset the impact of changes in domestic interest rates on credit variables. This also implies that EME credit conditions may become more vulnerable, particularly, to US monetary policy shocks. The results reported by Cettorelli and Goldberg (2012) suggest that a 100-basis point increase in the fed funds rate reduces foreign lending of US large commercial banks by as much as 2.2 %, implying a significant contractionary effect on EMEs .

Second, the expansion of dollar debt in EMEs implies that their domestic credit conditions are now very closely connected to the availability of dollar liquidity. This was, for instance, vividly demonstrated following the collapse of Lehman Brothers in 2008, which spread shock waves across the globe, causing large-scale dollar shortages and huge deleveraging pressures on EMEs (McGuire and von Peter 2009). As the shock transmitted to the FX swap markets, the cost of dollar funding escalated to very high levels, precipitating a broadly based tightening of credit conditions across EMEs.¹⁵ More generally, as argued by Borio et al. (2011), the sharp rise in dollar liabilities of EMEs over the past decade has meant that EME credit cycles have become more synchronized with the cycles in cross-border financing. In typical boom periods, cross-border credit tends to grow faster than growth in overall credit, with banks resorting to wholesale dollar funding markets to finance new asset growth.¹⁶ The process reverses itself with higher US interest rates, leading to large-scale unwinding of dollar borrowings and widespread credit slowdown in EMEs.¹⁷

Finally, to the extent that risk taking activities dominate, the credit channel is likely to become an important financial stability concern for many EMEs. Unfortunately, compared to the voluminous literature that exist on the bank lending channel, there is very little empirical work on the risk taking channel in EMEs, owing largely to the lack of detailed historical data on individual lenders and borrowers. That said, evidence based on aggregate credit data provides some useful guidance. For instance, using a cross-country panel model, Kohlscheen and Rungcharoenkitkul (2015) found that external factors such as the US dollar exchange rate and the implied US stock market volatility (VIX) have become more significant drivers of credit growth in EMEs after the 2008 crisis than they were before the crisis. Consistent with the risk taking channel of the exchange rate, their results suggest that a 10 % appreciation of EM currencies against the dollar is associated with about 85 basis points increase in credit growth in EMEs in the short run and 135 basis points increase in the long run .

4 Monetary Policy Transmission in India: An Empirical Assessment

In this section, we try to find out which of the channels of monetary policy are important in India. In addition to the three specific channels discussed in the previous section, we would also consider the asset price channel operating through equity prices. There have been some studies already examining the monetary policy transmission in India. For example, Al-Mashat (2003), using a quarterly structural vector error correction model (VECM) for the period of 1980–2002, found that interest rate and exchange rate channels were important, while due to the presence of directed lending regulations, evidence on the working of bank lending channel was weak. However, Pandit et al. (2006) found the existence of bank lending channel with small banks being more responsive to a policy shock. Singh and Kalirajan (2007) contended that the significance of policy interest rate substantially increased in India in the post 1990s reform period.¹⁸

In our analysis, we also try to answer the question whether US-specific shocks have an impact on Indian monetary conditions. That is, in addition to testing for the traditional channels of transmission we check whether US shocks directly transmit to India’s long-term bond yields and whether such shocks are met with policy reaction by the central bank, much similar to the enquiry by Miyajima et al. (2014).

4.1 Data and Methodology

We use a structural vector autoregressive (SVAR) framework and examine one channel at a time, like Khundrakpam and Jain (2012), keeping in mind that only a limited number of variables should be considered in order not to lose too many degrees of freedom. Therefore, we first estimate a baseline SVAR model and then augment the model by other channels. A way in which we differ with the previous studies on India is by taking US monetary policy shocks explicitly into consideration.

All data are quarterly, seasonally adjusted, and converted into logs (except the rate variables). Most of the variables were found to be integrated of order one and hence were first differenced to make them stationary.

We first run a baseline SVAR model without any of the above transmission channel variable. We use the following variables: 10-year US term premium (us_gov10_premia), real non-agricultural non-government GDP (ind_rgdp_nang) for output, wholesale price index (WPI, ind_wpi) for prices, and weighted average call rate (ind_call_rate) as policy rate. 10-year US term premium is used as a proxy for foreign monetary influence. Non-agricultural, non-government GDP is defined as total GDP excluding agriculture and allied activities, and "community, social and personal services.” We use this variant of GDP as it is expected to be more responsive to interest rate changes.

As for structural restrictions, of our domestic variables, weighted call rate responds contemporaneously to real GDP and WPI but not vice versa. Further, WPI is modeled as depending contemporaneously on real GDP. Finally, the US specific variables are exogenous to all domestic variables in the system. Therefore, the variables in the baseline model are ordered as {us_gov10_premia, ind_rgdp_nang, ind_wpi, ind_call_rate}. The baseline model thus gives us an idea of how the domestic interest rate channel works in terms of influencing output and inflation. Further, we can also check if US monetary policy has played any role in the evolution of Indian macroeconomic variables and policy interest rate setting.

Turning to augmenting our model with other transmission channels, we keep the baseline restrictions same and assume that the additional channel is contemporaneously affected by all the other variables included in the baseline model. Thus, the variables in the augmented model are ordered as {us_gov10_premia, ind_rgdp_nang, ind_wpi, ind_call_rate, "channel variable"}. The channel variables are non-food credit (ind_nbfoodcr) for the credit channel, BSE Sensex (ind_bsesensex) for the asset price channel, the real effective exchange rate (REER, ind_reer), and the nominal effective exchange rate (NEER, ind_neer) for the exchange rate channel and 10-year government bond yield (YLD_IND_10) for the bond price channel.

Additionally , in both the set of models, we use gross portfolio inflows as an exogenous variable to account for other global influences. We also include a dummy variable that takes value 1 in the peak crisis period (2008 Q1–2009 Q1) and zero otherwise. The analysis is done for the data from 1996 Q2–2014 Q4.

4.2 Results: The Baseline Model

We present the impulse responses from the baseline model in Graph 7 which shows the dynamic responses of output and inflation to one standard deviation shock in the call rate. Both output and inflation respond negatively to positive shocks in the call rate—as should be expected in models with sticky prices. We can further notice that in response to a call rate shock, the decline in GDP growth precedes the negative impact on inflation and the peak impact on inflation is also felt with a lag of one-quarter from the peak impact on GDP .

Graph. 7

Impulse responses from the baseline model

The remarkable negative response of output to shocks in the call rate may partly be due to the fact that we use the nonagricultural, nongovernment GDP as a measure of output, therefore, by definition, excluding less interest rate sensitive portion of GDP. However, this result is robust to using the aggregate real GDP, where the responses are still negative and significant, albeit smaller. Further, we also checked if these results are influenced by our identification strategy, causing contemporaneous correlation between output and the call rate. We, therefore, run the model without the restriction that the call rate responds contemporaneously to output (keeping all other restrictions the same as before) and we reach the same conclusion about the dynamic response of output to call rate shocks. In short, our results confirm the findings of others such as Aleem (2010), Khundrakpam and Jain (2012) that shocks to the call rate do have predictable effects on the economy .

For the US term premium, we do not find any evidence of it affecting the output or the call rate. Thus, over the period of the study, US monetary policy does not seem to have had any major influence on Indian monetary policy setting. To check for the importance of a direct long-term interest rate channel, we augmented the benchmark model with a bond price variable, represented by the yield on 10-year Government of India bond. Miyajima et al. (2014) in a very similar model specification found that the US long-term interest rate has remarkable influence on the long-term interest rates in the East Asian economies. We do not expect such dramatic results in the case of India for the reasons we discussed above, viz., relative insulation of Indian debt markets from international participation. The impulse responses for this model are presented in Graph 8. As expected, the 10-year Indian bond yields show no significant response to US term premium shock. Further, all the baseline results still hold.

Graph. 8

Impulse responses from the augmented model with Indian bond yield

Our results are consistent with the findings of Ghosh et al. (2017) in this volume that India has been able to keep its monetary policy independence from external forces intact due to its more cautious approach to bond market liberalization. Although India has progressively relaxed limits on foreign inflows into its domestic bond market in the recent years, the share of foreign ownership in the government bond market still remains very low. Another reason for incomplete arbitrage is limited capital account convertibility, which restricts borrowings by resident firms and households in the international debt markets. Our results complement Ghosh et al.’s analysis in another direction. They examine the impact of US monetary policy on capital flows to India and conclude that equity flows are more sensitive to global risk aversion but debt flows react more strongly to US interest rates. We investigate the same question by focussing on the price channel of transmission of US unconventional monetary policy, which is influenced by but does not necessarily depend on the quantity of capital flows (interest rates can move even without underlying changes in quantities of flows) .

4.3 Results: Augmented Model with Additional Transmission Channels

Through the augmented models we want to test whether and how short-term policy rate and US term premium shocks transmit to other channels of monetary transmission.

The impulse responses for models including transmission variables are presented in Graphs 9, 10, 11. The first is the credit channel which is expected to work on top of the direct interest rate channel. The efficacy of this channel obviously depends upon whether a significant number of borrowers are dependent on banks. Further, a higher US term premium can reduce credit growth by weakening the banks’ balance sheets (due to the expected higher non-performing loan ratio) and making them more risk averse. Graph 9 shows the relevant impulse responses. While results from the baseline model hold we find that a positive shock to the call rate leads to a decline in non-food credit. By contrast, the response of nonfood credit to US term premium shock is not significant.

Graph. 9

Impulse responses from the augmented model with credit channel

Graph. 10

Impulse responses from the augmented model with asset price channel

Graph. 11

a Impulse responses from the augmented model with exchange rate channel (REER). b Impulse Responses from the Augmented Model with Exchange Rate Channel (NEER)

Next, we turn to assessing the asset price channel. A tightening of monetary policy is expected to make equity, as an asset, less attractive compared with other alternative assets, such as bonds, leading to a fall in equity prices, and this may, through Tobin’s q, reduce investment. Also, a decline in the equity prices may reduce consumption demand through a net wealth effect on households. Similarly, a US term premium rise is expected to push asset prices down, for example, by leading to capital outflows. As a proxy for the asset price channel in India, we use the BSE SENSEX—the most popular index of Indian equity prices. The relevant impulse responses are shown in Graph 10. We find again that the results of the baseline model hold. Also clearly, a positive shock to the policy rate leads to a decline in equity prices, which peaks after two quarters following the shock. However, we do not see the US term premium shock having any significant effect on asset prices .

Finally , we look at the exchange rate channel. We consider both the REER and the NEER for the exchange rate channel. Graph 11 presents relevant impulse responses when REER is taken as the transmission channel. Here we find a result that is contrary to what we expect. While the baseline results hold true again, we find a positive shock to the policy rate that leads to a real exchange rate depreciation. Khundrakpam and Jain (2012) explain this diversion from the UIP by contending that interest rate differentials do not play an important role in the exchange rate determination in India. As discussed before, a part of the explanation is that the most interest sensitive part of capital flows, i.e., the debt component, is restricted in India, reducing the currency appreciation effect of a higher interest rate. Another could be that a tighter monetary policy may actually reduce equity flows by weakening growth prospects. Thus, it is possible that, on a net basis, a policy rate shock leads to a decline in net capital inflows and a currency depreciation. We find a similar result when we take the NEER as a transmission channel. In case of a US term premium shock, we find that it has no significant effect on the REER or NEER for the reasons mentioned above.

To sum up, we find that the interest rate is an important channel of monetary policy transmission in India as it has significant effect on output and inflation. Among, the other channels, both the asset price channel and credit channel are active while the bond price and exchange rate channels in India do not work in the expected way mainly due to its capital account policies. Keeping Miyajima et al. (2014) results in mind, thus provide indirect evidence that external monetary policy shocks are transmitted to the domestic financial systems through globally integrated debt markets which reduces barriers to international arbitrage and equalizes long-term rates across economies. In other words, the “impossibility trinity” holds, i.e., an open debt market limits the sphere of influence of domestic monetary policy on the economy. In what follows we use a simple reduced form model to demonstrate the challenges facing the monetary authority .

5 Interest Rate Setting in Globalized Debt Markets

5.1 The Closed Economy Case

In this section, we present a simple monetary model to illustrate how a globalized debt market might complicate central bank’s response. Following Genberg (2008), the analytical model comprises four equations. These equations are written in the simplest form and the lagged values are suppressed but they may nevertheless be important. We also assume a frictionless economy, so imbalances such as currency mismatches and credit market imperfections do not play a role. We first start with a closed economy model

$$ {{\pi_{\rm{t}}}} = {{\upalpha}} \,{{\rm Y}}\,{\rm{gap}}_{\rm{t}} + {{\upvarepsilon}}_{{\rm{t}}}^{{{\uppi}}} $$

(6)

$$ {\text{Ygap}}_{\text{t}} =\upbeta_{1} \left[ {{\text{i}}_{\text{t}}^{\text{m}} -\uppi_{\text{t}} } \right] +\upbeta_{2} {\text{U}}_{\text{t}}^{\text{c}} +\upvarepsilon_{t}^{\text{Ygap}} $$

(7)

$$ {\text{U}}_{\text{t}}^{\text{c}} = {\text{P}}_{\text{t}}^{\text{c}} [{\text{E}}{\mathbf{ }}\{ {\text{i}}_{\text{t}}^{\text{m}} -\uppi_{\text{t}}^{\text{c}} \} +\updelta] $$

(8)

Equation 6 is a simple version of the Phillip curve where inflation $(\uppi_{\text{t}} )$ depends on output gap $( {\text{Ygap}}_{\text{t}} )$. The error term ${\varvec{\upvarepsilon}}_{{\mathbf{t}}}^{{\varvec{\uppi}}}$ captures the omitted factors and a white noise component. Equation 7 defines the IS relation with output gap being dependent on the short-term nominal market interest rate $({\mathbf{i}}_{{\mathbf{t}}}^{{\mathbf{m}}} )$ net of inflation $({\varvec{\uppi}}_{{\mathbf{t}}} )$ and the expected user cost of capital $({\mathbf{U}}_{{\mathbf{t}}}^{{\mathbf{c}}} )$. The last term again is the error and has the usual interpretation as above. Equation 8 is the user cost of capital equation, hence identical to Eq. (1).

By substituting the expected user cost expression in the IS equation specified above, we have

$$ {\text{Ygap}}_{\text{t}} =\upbeta_{1} \left[ {{\text{i}}_{\text{t}}^{\text{m}} -\uppi_{\text{t}} } \right] +\upbeta_{2} {\text{P}}_{\text{t}}^{\text{c}} \left[ {{\text{E }}\left\{ {{\text{i}}_{\text{t}}^{\text{m}} -\uppi_{\text{t}}^{\text{c}} } \right\} +\updelta} \right] +\upvarepsilon_{t}^{\text{Ygap}} $$

(9)

Next we define the central bank reaction function in Eq. (10), which links the policy rate $({\rm{i}}_{{\rm{t}}}^{{\rm{p}}} )$ to the output gap and deviation of actual inflation $({{\uppi}}_{{\rm{t}}} )$ from the target $({{\uppi}}^{{\rm{T}}} )$:

$$ {\text{i}}_{\text{t}}^{\text{p}} =\upgamma_{0} +\upgamma_{1} \left[ {\uppi_{\text{t}} -\uppi^{\text{T}} } \right] +\upgamma_{2} \,{\text{Ygap}}_{\rm{t}} +\upvarepsilon_{\text{t}}^{{i^{p} }} $$

(10)

This is basically a form of the Taylor rule (Taylor 2003). Note, however, that we have an additional term $({{\upgamma}}_{0} )$ in the interest rate rule usually assume to be a constant to capture other factors that have become important for central banks in EMEs.

Our discussion in the previous sections suggests that the size and the form of financial intermediation are crucial for transmission of monetary policy shocks. Most notably, firms have been able to access bond markets to finance investment. In a closed economy framework, this could, for instance, mean that the market borrowing rate $({\rm{i}}_{{\rm{t}}}^{{\rm{m}}} )$ is now more tightly linked to the policy interest rate $({\rm{i}}_{{\rm{t}}}^{{\rm{p}}} )$.

We, therefore, need a relationship between $({\rm{i}}_{{\rm{t}}}^{{\rm{p}}} )$ and $({\rm{i}}_{{\rm{t}}}^{{\rm{m}}} )$ that takes into account several institutional factors that create a wedge between the policy interest rate and the market interest rate which we denote by ${{X}}_{{\rm{t}}}$. It may also include the effects of pricing power of banks and structural impediments to interest rate setting in the banking system. It can also be thought to be capturing general level of risk aversion or perceived credit risk in the economy:

$$ {\rm{i}}_{{\rm{t}}}^{{\rm{m}}} = {{\uptheta}}_{1} {\rm{i}}_{{\rm{t}}}^{{\rm{p}}} + {{\uptheta}}_{2} {\rm{X}}_{{\rm{t}}} + {{\upvarepsilon}}_{{\rm{t}}}^{{\rm{i}}^{{\rm{m}}} } $$

(11)

In order to characterize the static equilibrium market interest rate $({\rm{i}}_{{\rm{t}}}^{{{\rm{m}}^{*} }} )$ and equilibrium policy rate $({\rm{i}}_{{\rm{t}}}^{{{\rm{p}}^{*} }} )$, we solve for the market interest rate and policy interest rate in Eqs. (9) and (10) when output gap and inflation gaps are zero. We therefore have

$$ {\rm{i}}_{{\rm{t}}}^{{{\rm{m}}^{*} }} = {{\uppi}}^{{\rm{T}}} - \frac{{{{\upbeta}}_{2} }}{{{{\upbeta}}_{1} }} {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} \left[ {{\rm{E}} \left\{ {{\rm{i}}_{{\rm{t}}}^{{\rm{m}}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} } \right\} + {{\updelta}}} \right] + \frac{1}{{{{\upbeta}}_{1} }}{{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{Ygap}}}} $$

(12)

$$ {\rm{i}}_{{\rm{t}}}^{{{\rm{p}}^{*} }} = {{\upgamma}}_{0} + {{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{p}}} }} $$

(13)

Note, again the error terms have the usual meaning as discussed above. Now, we are ready to find an expression for ${{\upgamma}}_{0}$ and using that we augment the traditional Taylor rule . Thus, by substituting Eqs. (12) and (13) in Eq. (10), we have

$$ {{\upgamma}}_{0} = \frac{1}{{{{\uptheta}}_{1} }}({{\uppi}}^{{\rm{T}}} - \frac{{{{\upbeta}}_{2} }}{{{{\upbeta}}_{1} }} {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} \left[ {{\rm{E}} \left\{ {{\rm{i}}_{{\rm{t}}}^{{\rm{m}}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} } \right\} + {{\updelta}}} \right]) - \frac{{{{\uptheta}}_{2} }}{{{{\uptheta}}_{1} }}{\rm{X}}_{{\rm{t}}} + \frac{1}{{{{\uptheta}}_{1} }}\frac{1}{{{{\upbeta}}_{1} }}{{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{Ygap}}}} - \frac{1}{{{{\uptheta}}_{1} }}{{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{m}}} }} - {{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{p}}} }} $$

where $\frac{1}{{{{\uptheta}}_{1} }}\frac{1}{{{{\upbeta}}_{1} }}{{\upvarepsilon}}_{{{t}}}^{{{{Ygap}}}} - \frac{1}{{{{\uptheta}}_{1} }}{{\upvarepsilon}}_{{{t}}}^{{{{i}}^{{{m}}} }} - {{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{p}}} }}$ are a bunch of errors which we denote by ${{\upvarphi_{t}}}$.

Finally, our simplistic but augmented monetary policy reaction function can be written as

$$ {\rm{i}}_{{\rm{t}}}^{{\mathbf{p}}} = \frac{1}{{{{\uptheta}}_{1} }}({{\uppi}}^{{\mathbf{T}}} - \frac{{{{\upbeta}}_{2} }}{{{{\upbeta}}_{1} }} {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} \left[ {{\rm{E}} \left\{ {{\rm{i}}_{{\rm{t}}}^{{\rm{m}}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} } \right\} + {{\updelta}}} \right] - {{\uptheta}}_{2} {\rm{X}}_{{\rm{t}}} ) + {{\upgamma}}_{1} \left[ {{{\uppi}}_{{\rm{t}}} - {{\uppi}}^{{\rm{T}}} } \right] + {{\upgamma}}_{2} \,{\rm{Ygapt}} + {\rm{\upvarphi_{t}}} + {{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{p}}} }} $$

(14)

Ignoring the two error terms, the last two terms are the usual Taylor rule arguments. The first term explicitly brings into the policy rule, the expected user cost, and financial intermediation factors. This relation will be used as baseline when we open up the economy to international bond flows .

Two main implications of the closed economy augmented Taylor rule are worth noting. First, as discussed by Genberg (2008), it predicts how financial intermediation can affect the "neutral interest rate" to stabilize the economy. One example, discussed in Sect. 3, is the case of increased bond financing leading to lower intermediation spreads. This is akin to monetary easing and thus to maintain neutral monetary stance, the central bank must respond by increasing the policy rate. In contrast, an increase in perceived credit risk, unrelated to fundamentals of the economy, or higher intermediation spreads resulting from attenuation of asymmetric information problem through the bond market, will mean monetary conditions have tightened, thus requiring a lower policy rate.

Second, our augmented central bank reaction function also illustrates how market expectations of the interest rate and asset prices play an important role in the design of appropriate interest rate response, particularly when they deviate from those desired by the central bank. This is likely to affect the size and the speed of the response of aggregate demand to changes in the interest rate as well as other shocks driving these expectations. For instance, a cut in the policy rate may be accompanied by unexpected buoyancy in house prices, leading to faster than expected transmission of policy shocks to the housing market, which may not be welcome by the central bank from the view point of financial stability. The same phenomenon could also be driven by house price inflation completely unrelated to monetary policy (demographic shocks), which increases the perceived pay-off to home buyers from housing investment, driving spending and inflation away from target.

The main point to note is that the conventional Taylor rule still serves a useful purpose in stabilizing the economy. To the extent that changes in domestic financial intermediation are gradual, central banks can prevent major risks to monetary and financial stability by appropriately adjusting their policy response. Precisely for this reason, in many countries, monetary authorities regularly monitor a wide range of real and financial market indicators to detect and address some of these risks .

5.2 The Open Economy Case

Using the above framework we can also bring in the open economy considerations into the picture. Consider the expected user cost equation in an open economy which is identical to Eq. (3):

$$ {\rm{U}}_{{\rm{t}}}^{{\rm{c}}} = {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} [{\rm{E}}\,\{ {\rm{i}}_{{\rm{t}}}^{{{\rm{us}}}} + {{\Delta}}{\rm{e}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} \} + ( {\rm{q}}^{{{\rm{us}}}} - {\rm{q}}^{{\rm{d}}} ) + {{\uprho}} + {\rm{\updelta}}] $$

We want to arrive at the open economy interest rate rule as we did in case of the closed economy. Equation (15) provides such an expression:

$$ \begin{aligned} {\rm{i}}_{{\rm{t}}}^{{\rm{p}}} & = \frac{1}{{{{\uptheta}}_{1} }}({{\uppi}}^{{\rm{T}}} - \frac{{{{\upbeta}}_{2} }}{{{{\upbeta}}_{1} }} {\rm{P}}_{{\rm{t}}}^{{\rm{c}}} [{\rm{E}}\,\{ {\rm{i}}_{{\rm{t}}}^{{{\rm{us}}}} + {{\Delta}}{\rm{e}} - {{\uppi}}_{{\rm{t}}}^{{\rm{c}}} \} + ( {\rm{q}}^{{{\rm{us}}}} - {\rm{q}}^{{\rm{d}}} ) \\ & \quad + {{\uprho}} + {{\updelta}}] - {{\uptheta}}_{2} {\rm{X}}_{{\rm{t}}} ) + {{\upgamma}}_{1} \left[{{{\uppi}}_{{\rm{t}}} - {{\uppi}}^{{\rm{T}}} } \right] + {{\upgamma}}_{2} \,{\rm{Ygapt}} + {{\upvarphi }} + {{\upvarepsilon}}_{{\rm{t}}}^{{{\rm{i}}^{{\rm{p}}} }} \\ \end{aligned} $$

(15)

The open economy interest rate rule clearly demonstrates how foreign factors may have a bearing on monetary policy setting. According to Eq. (15), the neutral policy interest rate, in a globalized environment, will have to be set taking into account domestic and US term premia, currency risk premia and expected exchange rate movements . We can now think of several scenarios where policy rate has to be maneuvred independent of core domestic objectives. Take, for instance, the case of a lower US term premium that results from policy actions such as large-scale quantitative easing by the fed. This leads to rapid exchange rate appreciation, which reduces the cost of credit in the shortrun. Note that this leads to an easing of domestic monetary conditions, requiring a higher domestic interest rate to stabilize inflation. Such a strategy is especially problematic when the domestic fundamentals require an opposite action, and would be unsustainable if higher interest rates encouraged more capital inflows. A reverse scenario may quickly develop if the US term premium starts to rise, for instance, that happened during the "taper tantrum" episode, leading to a sudden tightening of monetary conditions.

Equation (15) also predicts the neutral rate in more direct cases of the US Federal Reserve’s forward guidance such as its intention to maintain a zero fed funds rate into the future. Such guidance directly changes the expected path of future US short-term rates and hence credit costs facing EME borrowers. Again, it is difficult to envisage the central bank playing by Eq. (15). In practice, the dilemma could be more complicated because exchange rate may appreciate too fast and too soon and that such appreciation may trigger an unwelcome credit boom, raising risks to financial stability.

In short, with globalization of debt markets, monetary policy conduct through the short-term interest rate becomes a much more complicated affair. A simple Taylor rule is unlikely to be sufficient in stabilizing the economy against external monetary shocks, requiring monetary authorities to depend on multiple instruments to balance domestic and external objectives (e.g., Obstfeld (2015). India’s approach has suggested that the degree of capital account flexibility—and therefore the choice of monetary policy regime—plays an important role in determining the impact of external shocks .

6 Conclusion

The objective of this paper was to review changes in the monetary transmission mechanism in EMEs following several major changes to financial intermediation over the past decade. It is by now fairly obvious that the globalization of debt markets, together with a sharp decline in global long-term interest rate and accumulation of large dollar debt by EME nonfinancial corporations, have complicated the transmission mechanism of monetary policy in many economies. One well-known consequence of reduced barriers to international arbitrage is that domestic asset prices cannot deviate too much from international asset prices. The analysis in this paper suggested that these changes have affected all three major channels of monetary policy creating new tensions for monetary and financial stability policies. However, as pointed out by Ghosh et al. (2017) in this volume, India has well-navigated recent large global shocks because of its cautious approach to bond market liberalization. Our results suggested that monetary policy continues to play a significant role in the macroeconomic evolution of the Indian economy.

A key question is the extent to which the recent changes to the transmission mechanism affect interest rate setting, especially in small open economies. Our results suggest that the traditional Taylor rule can still be a reasonable guide to monetary policy in relatively closed economies subject to gradual changes in financial intermediation. In this case, the neutral interest rate could be adjusted to prevent major inflation risk due to potential changes in the response of aggregate demand to interest rate. The challenges are more complicated in globally integrated debt markets. As discussed by Obstfeld (2015), Agenor and Pereira de Silva (2013), a single interest rate instrument is unlikely to be a satisfactory solution in most cases, requiring the central bank to use other instruments (such as foreign exchange intervention and macro-prudential tools) to reduce risks to price and financial stability.