Printer Friendly

The Impact of the US Unconventional Monetary Policy and Its Normalization in the Philippines: A Capital Flow Perspective.

1. Introduction

Capital surges and subsequent reversals of greater magnitude have been observed in emerging market economies (EMEs) stemming from spillovers from US monetary policy aimed at mitigating the effects of the Global Financial Crisis (GFC) that started in mid-2007. Throughout this period, the US Federal Reserve (US Fed) has used unconventional monetary policy tools (such as setting its interest rate to zero lower-bound and implementing quantitative easing (QE) measures) to revive the growth of the US economy. In the Philippines, surges in capital flows became apparent starting in 2010. However, reversals of greater magnitude have since been observed in several EMEs, including the Philippines, during the following episodes: first, when the former US Fed Chair Bemanke first mentioned the possibility of tapering in May 2013; second, during the first announcement of the US Fed that it would start the tapering process in December 2013, continuing after the end of QE in October 2014; and third, during the US Fed pronouncement of the first Fed rate hike in December 2015. Extreme movements in capital flows in response to the US monetary policy actions have implications on domestic capital management measures. In particular, with the expected resumption of the US Fed rate hike as macroeconomic conditions strengthen, policymakers in EMEs need to stand ready to counter extreme capital flow volatilities and build cushions to ensure resilience.

While studies have shown that previous episodes of EME currency crises (for instance, Latin America in 1982, Mexico in 1994, and Asia in 1997) were characterized by surges in capital flows that were subsequently followed by reversals or "sudden stops", some recent studies have narrowed down the effects of the US unconventional monetary policy and its normalization on the behaviour of specific capital flows. These studies stress that different types of capital flows could be more sensitive to changes in monetary policy and that some capital flows may have different implications for the financial stability of recipient countries. For instance, Suh and Koo (2016), taking into account both "push" (i.e., global) and "pull" (i.e., domestic) factors in the analysis, find that the effects of unconventional monetary policy and its normalization on capital flows in EMEs are sizeable on average, especially for capital flows via both cross-border borrowing and bonds channels, but the impact is insignificant via the equity channel. (1) Similarly, Lim, Mohapatra and Stocker (2014) show that, among the different types of capital flows in EMEs, portfolio bond flows tend to be more sensitive to QE. The same study finds that QE withdrawal could lead to a contraction in the capital inflows-to-GDP ratio among EMEs of 0.6 per cent. They add that foreign direct investment (FDI) is largely insensitive to QE and policy normalization. Meanwhile, De Gregorio (2014) stresses the importance of debt portfolio flows and banking flows (i.e., bank and money market flows) compared to equity flows. He notes that equity flows tend to carry very few risks and are largely insensitive to changes in monetary policy, while the two former flows tend to be riskier as they carry maturity and currency risks. Furthermore, Sun (2015) shows that the expansion of global liquidity due to QE has a pronounced cyclical nature subject to occasional adverse shocks and that tightening of global financial conditions carries significant risks for the Association of Southeast Asian Nations Five (ASEAN-5). (2)

Given the aforementioned findings, this paper tries to explore the case of the Philippines as an example of a small, open economy. Specifically, this study seeks to determine the impact of the US unconventional monetary policy and its normalization on the dynamics of capital inflows into the Philippines, particularly the relatively volatile flows, namely: portfolio equity; portfolio debt securities; and other investments (mostly composed of bank and money market flows).

Existing research conducted for the Philippines on the said topic has focused largely on the effects of the US unconventional monetary policy, but little on the effects of policy normalization. Moreover, these studies have only examined the behaviour of total capital inflows, but not the dynamics of specific types of the volatile inflows (see, for example, Fernandez and Kalaw 2016). Therefore, this research attempts to fill the gap in the literature by evaluating the impact of QE and the US monetary policy normalization on portfolio equity, portfolio debt securities, and bank and money market inflows into the Philippines. The analysis will measure the magnitude and persistence of the aforementioned monetary policy shocks on specific types of capital inflows, using both "push" and "pull" factors in an SVAR model.

This paper is structured as follows. The next section discusses capital flows trends and policies implemented in the Philippines. The third section describes the relevant theoretical framework, empirical background, as well as the model and data set used in the study. The subsequent section presents the results and evaluates the plausibility of such results in light of earlier studies. The fifth section elaborates on the robustness check that this paper utilized, and the last section concludes.

2. Capital Flows in the Philippines: Trends and Policies (3)

2.1 Drivers of Capital Flows in the Philippines

The movements in capital inflows into the country in the last three decades were influenced by both domestic as well as global developments. For instance, the capital surges experienced by the Philippines in years prior to the Asian Financial Crisis (AFC) were driven by domestic reforms that liberalized the country's foreign exchange (FX) system. These were complemented by efforts pursued by the government to attract and promote FDI. Meanwhile, the major cause of capital reversal from 1997 to 2004 was the 1997 AFC, which affected the Philippines through contagion from its Asian neighbours. Similarly, the relative improvement in macroeconomic conditions, both in domestic and global settings influenced the rapid growth in capital inflows into the country between 2005 and 2007. For the period lasting from 2008 to 2015, even with some reversals in the relatively volatile capital inflows, the strength of FDI has supported the growth of private capital inflows during the period (see PHDI in Figure l). (4)

Table 1 indicates the magnitude of the country's gross private capital inflows-to-nominal GDP. As shown in the table, the country's gross private capital inflows-to-nominal GDP has been dominated by FDI, which accounts for around 2 per cent. Meanwhile, of the volatile capital inflows, portfolio debt securities, and bank and money market inflows comprise relatively larger shares compared to portfolio equity. A salient feature of the table is the sustained growth of the country's FDI from 2010 to present, indicating that FDI has been a consistent driver of capital inflows into the country in recent years.

The GFC and the subsequent monetary policy responses in advanced economies have also influenced capital flows into the Philippines. In 2008, capital flows, particularly the relatively volatile flows, turned negative as the crisis unfolded (see PHPE, PHPD and PHBM in Figure 1). Monetary authorities in advanced economies adopted aggressive expansionary monetary policy measures to address the GFC by decreasing interest rates swiftly. When interest rates reached the zero lower-bound, they introduced unconventional monetary policies beginning in late 2008 to prop up growth (i.e., the US from 2008 to 2012, Europe since 2009, Japan since 2010, and the United Kingdom from 2009 to 2012). (5) The conventional and unconventional monetary measures were implemented to decrease financing costs. However, as yields decreased in advanced economies, investors searched for higher yields elsewhere. A higher global demand for EME assets by international investors generated increased capital inflows into EMEs. As shown in Figure 2, surges in capital flows in the Philippines, especially portfolio debt (PHPD), and bank and money market (PHBM) are noticeable amidst volatility during the QE period.

It is apparent that the monetary policy actions in advanced economies, particularly those in the US, have left a small, open economy like the Philippines vulnerable to increased capital flow volatility. Figures 3, 4, and 5 show the trends of these individual capital inflows into the Philippines during the pre-QE period (1999-2007), at the height of the QE implementation (starting late 2008), the start of tapering (late 2013), and the start of policy normalization process (late 2015). During the QE period, the Philippines experienced capital surges that heightened fluctuations of the inherently volatile capital inflows such as the portfolio equity, portfolio debt securities, and bank and money market. Among these types of capital, volatility was more pronounced in portfolio debt securities, and bank and money market.

In addition to external factors, the Philippines attracted large FX inflows because of strong economic growth prospects. Since 2010, the Philippines has established a reputation for its solid macroeconomic fundamentals, including high growth levels, stable and within-target inflation, sound external payments position, ample international reserves, manageable external debt, good fiscal performance, and favourable demographic factors. Such bright prospects have led to successive sovereign credit rating upgrades to investment grade by major credit rating agencies such as Standard & Poors, Moody's Investors Service, and Fitch Ratings from 2010 to 2015. These combined impact of "push" and "pull" factors has significantly influenced the direction of capital flows during the period.

2.2 Policies Implemented to Manage Capital Flows

2.2.1 Pre-US QE Period. Prior to the AFC, the Philippines experienced significant surges in capital flows during the 1992-96 period. To cope with the capital flow surges and the consequent adverse effects, the Bangko Sentral ng Pilipinas (or the Central Bank of the Philippines, henceforth, BSP) implemented several policy measures, including the liberalization of the FX regulatory framework to stem the appreciation of the peso and avoid substantial loss of competitiveness. Other measures included accelerating the BSP's prepayment of some of its external debt (around US$1.4 billion) to reduce the supply of FX. Rules on capital outflows were also relaxed by increasing allowable outward investments that could be sourced from the banking system--from US$1 million to US$6 million per investor per year. The BSP also lifted the restrictions on the repatriation of foreign investments.

Meanwhile, from 1997 to 2004, the Philippine economy had to cope with capital reversals. As the country experienced large capital flow reversals and to avoid extreme exchange rate depreciation, the BSP devised a delicate balancing act of limiting currency depreciation and allowing a temporary rise in interest rate. The BSP also undertook an aggressive and comprehensive reform process of the domestic financial system. (6) Similarly, reforms such as improvement in transparency in monitoring capital flows through Early Warning Systems (EWS) for currency crises and bank failure were introduced.

From 2005 to early 2007, capital inflows into the Philippines increased gradually, largely due to improvements in both the domestic and global outlook. To deal with large capital inflows, including sustained FX inflows from the Overseas Filipino (OF) remittances and exports, the BSP implemented reforms in the FX regulatory framework. (7) Subsequently, reforms that focused on promoting greater integration with the international capital markets and risk diversification, and streamlining the documentation and reporting requirements on the sale of FX by banks were also implemented. (8) Meanwhile, liquidity management measures were pursued to help prevent inflationary pressures that could build up over the medium term, mainly as a result of rapid money supply growth. (9) The BSP also increased its reserve buffer to counter sudden capital flow reversal and the possible unwelcome changes in the value of the domestic currency. Also, prepayment of external loans was accelerated. (10)

2.2.2 US QE Period. During the GFC and after the implementation of QE by the US in November 2008, capital inflows into the Philippines started to show strength, especially at the start of 2009, and surged in 2010. The main drivers of these surges were capital inflows into portfolio debt securities, and bank and money market. Meanwhile, portfolio equity inflows and FDI remained stable. This trend continued in 2011 and 2012 (as previously shown in Figure 2). As a response to the surges primarily caused by the US unconventional monetary policy, the BSP continued to adopt existing policy measures, including: (1) improved monitoring of FX inflows; (2) exchange rate flexibility; (3) reserve accumulation; (4) enhanced monitoring of external borrowings; (5) liquidity management; (6) FX regulatory reforms; (7) capital market deepening; (8) foreign borrowings prepayment; and (9) monetary policy calibration. The BSP also adopted macro-prudential policy instruments to stabilize the financial system and influence directly the credit cycle. (11)

The surges in capital flows were temporarily disrupted during the taper tantrum in May 2013, when significant reversals were observed. However, the impact was tempered by the increase in FDI inflows that gave a boost to the growth prospects of the Philippine economy. Sustained, strong and non-inflationary growth marked the Philippine economy in 2013. Real GDP expanded by 7.2 per cent, surpassing market expectations. Domestic business sentiment remained favourable and helped support spending and investment decisions. At the same time, the country obtained investment grade status from all the three major credit rating agencies - Fitch Ratings, Standard & Poor's, and Moody's Investors Service.

2.2.3 Post-US QE Period. Monetary authorities in the Philippines remain vigilant as uncertainty surrounding the timing and magnitude of the hike in the US policy rate could affect debt and interest levels worldwide. The QE tapering started in December 2013 and subsequently ended in October 2014, shifting market expectation to the US policy normalization, and eventually triggering the reversal of capital flows. In December 2015, following the first Fed rate hike, uncertainty regarding the subsequent interest rate adjustments by the US Fed has contributed to increased financial market volatility in EMEs, including the Philippines.

While continuing its existing policies on managing capital flows, the BSP has adopted enhanced measures to safeguard financial stability. In particular, the BSP implemented the Real Estate Stress Test (REST) (12) to manage banks' exposure to the real estate sector. In addition, the BSP fine-tuned its guidelines on credit risk management framework and governance for financial institutions to mitigate credit concentration risks. (13) It also strengthened cooperation in managing capital flows. Unlike the case in other countries, the Financial Sector Stability Council (FSSC) in the Philippines is led by the BSP, with inter-agency policy coordination to support business and financial growth and safeguard financial stability. (14)

3. Analytical Framework

Empirical studies show that QE and the subsequent monetary policy normalization can trigger capital volatility in EMEs like the Philippines, because of which these US Fed policies are regarded as push factors of capital flows to EMEs. Suh and Koo (2016), for example, show that monetary policy normalization can lead to capital flow reversals from EMEs back to advanced economies from all types of capital, including borrowings, bonds, and equity flows. It is important to note that, in the case of EMEs, the adverse effects of capital reversal are larger than the boost received from capital inflows during QE (Koepke 2018). Dahlhaus and Vasishtha (2014) estimated that US monetary policy normalization decreases the capital flows-to-GDP ratio of EMEs by 0.5 per cent on impact and 1.2 per cent after three months. Specific results for the Philippines estimate that on impact, the US monetary policy normalization decreases capital flows-to-GDP by 0.8 per cent, relatively lower compared to the impact on other ASEAN countries like Malaysia (1.2 per cent) and Thailand (1.0 per cent), but higher than Indonesia (0.6 per cent) (Dahlhaus and Vasishtha 2014). At present, no certain consensus is reached on how large the impact of the US unconventional monetary policy and its normalization on capital flows in EMEs is.

Examining different flows is important since the effect of US unconventional monetary policy can differ depending on the type of capital flow. Between bond and equity flows, the impact on bond flows is more pronounced (Dahlhaus and Vasishtha 2014; Suh and Koo 2016). Suh and Koo (2016) estimate that changes in cross-border borrowings associated with the US unconventional monetary policy range from a contraction of 64 per cent to an increase of 351 per cent (or 32 per cent on average). In addition, they note that the negative relationship between unconventional monetary policy and cross-border borrowings of EMEs implies that factors other than the US unconventional monetary policy may have affected the capital inflows into EMEs in the opposite direction. In contrast, equity flows saw a slightly positive effect or remained unaffected by the US unconventional monetary policy. Koepke (2018) argues that bond flows react strongly to a surprise change to the US Fed funds rate when compared to equity flows because bond prices are closely related to interest rates than are stock prices.

This paper investigates the impact of the US unconventional monetary policy and its normalization on selective volatile flows represented by the following variables of interest (henceforth, [VI.sub.j]):

1. Philippine portfolio equity inflows scaled as a percentage of nominal GDP (SPHPE);

2. Philippine portfolio debt securities inflows scaled as a percentage of nominal GDP (SPHPD); and

3. Philippine bank and money market inflows scaled as a percentage of nominal GDP (SPHBM).

The US unconventional monetary policy and its normalization can have an impact on capital inflows of a small, open economy through three main channels: first, portfolio balance channel--suggests that the purchase made through QE reduces the supply of US assets to private investors, and therefore demand for substitute assets (i.e., EME assets) increases as investors adopt riskier behaviour in search for higher yields (Chen et al. 2013); second, signalling channel--underscores the importance of the US Fed actions as a way of providing information on the current state of the US economy, which consequently influences the appetite of investors as reflected in their investment decisions (Bauer and Rudebusch 2013); and third, liquidity channel--indicates that QE could improve market functioning and decrease liquidity premia. This, in turn, encourages investors away from US bonds to non-US bonds (Fratzscher, Lo Duca, and Straub 2013). A simplified representation of the aforementioned transmission channels is shown in Figure 6.

This study tests the impact of global or "push" factors and their corresponding channel, using the following: (15)

1. A variable representing the US unconventional monetary policy or level of liquidity in the US as proxied by M2 or broad money (USLIQ). This variable captures the liquidity and portfolio balance channels;

2. Variables representing the US policy normalization: (1) expected US policy rate path as proxied by fed funds futures rate (USFUT) to capture the change in expectations, part of the signalling channel; and (2) a dummy variable is also incorporated to capture a surprise change in the US monetary policy through announcement (i.e., taper tantrum, start of QE tapering, end of QE, and first Fed rate hike) (USSIG). The study examines the impact of a surprise change on the timing of the implementation of the policy normalization. However, there is no consensus yet among empirical studies about the impact of the timing. Koepke (2018) shows that the effect of monetary policy on EMEs capital inflows may also be due to the element of surprise. On the other hand, Lim, Mohapatra and Stocker (2014) show that regardless of whether normalization occurs gradually or rapidly, on average, the gross capital flows-to-GDP ratio increased by 0.08 per cent during QE and decreased by 0.6 per cent when policy normalization continued until the end of 2016.

3. This study also incorporates a variable representing change in market sentiments (i.e., business or household) represented by the US Chicago Board Options Exchange volatility index (USVIX) to capture the confidence channel effects, which is also part of the signalling channel.

Although this paper primarily investigates the impact of "push" factors to capital flows, this model also integrates "pull" factors as control variables. Ignoring the latter can lead to underestimation or overestimation of results as some macroeconomic fundamentals or idiosyncratic vulnerabilities of each country may not be accounted for (Lim, Mohapatra and Stocker (2014) The impact of the monetary policy normalization on EMEs may depend on each country's credit-to-GDP ratio dispersion, international debt securities, and external loan and deposits exposures (Sun 2015), as well as on whether a country is a recipient of large capital inflows (Suh and Koo 2016). In an empirical study of ASEAN-5 economies, Sun (2015) finds that economies also differ in terms of risks exposure. Indonesia and the Philippines are found to be at a low level, Malaysia and Thailand at a middle level, and Singapore at a high level. Thus, some studies consider both "push" and "pull" factors in analysing the impact of the US unconventional monetary policy and its normalization on capital flows to EMEs.

To control for the potential domestic factors affecting capital flows, the following domestic or "pull" factors are added in the SVAR model: (16)

1. Philippine short-term interest rate (PHSTR);

2. Philippine real effective exchange rate (PHRER); and

3. Philippine industrial production index (PHIPI).

The study uses publicly available monthly data from January 2008 to March 2016. (17) The description of variables can be found in Appendix A.

3.1 The SVAR Model

The study evaluates the impact of the US unconventional monetary policy and its normalization on specific [VI.sub.i] namely: (1) SPHPE; (2) SPHPD; and (3) SPHBM using an SVAR model. Global or "push" factors and domestic or "pull" factors are interacted with specific [VI.sub.i] to see how these inflows behave given the aforementioned monetary policy changes. USLIQ, USSIG, USFUT and USVIX are included to capture the QE measures, the US Fed's announcement of a surprise change in monetary policy (i.e., taper tantrum, start of tapering, end of QE, and initial Fed rate hike), expectations of future path of fed funds rate, and the US market volatility; respectively. Meanwhile, PHSTR, PHRER and PHIPI are included to reflect the macroeconomic fundamentals (i.e., interest rate, real effective exchange rate, and industrial production index) of the Philippines. Thus, this study generates three models of an eight-variable SVAR. Each model includes one variable of interest along with all the identified "push" and "pull" variables identified for the model. Within this framework, specific capital inflows can be modelled as follows:

[mathematical expression not reproducible] (1)

The use of SVAR model allows to recover economic shocks by imposing minimum assumptions compared to other large models (e.g., dynamic stochastic general equilibrium (DSGE) models), and it captures the dynamics of unexpected shocks (Rummel 2015; Gottschalk 2001). However, both studies note that one drawback of the SVAR methodology is its low dimensionality, which implies that the assumption that the underlying shocks are orthogonal is likely to be fairly restrictive. Nevertheless, the same studies emphasize that the SVAR methodology continues to be the workhorse of empirical macroeconomics and finance.

To assess the impact of the US unconventional monetary policy and its normalization on [VI.sub.j], where i = (1) SPHPE, (2) SPHPD, and (3) SPHBM, the identification methodology of SVAR described by Amisano and Giannini (1997), the AB-model, is applied. Hence, adopting that of Rummel (2015), the underlying structural system of equations is of the form:

[mathematical expression not reproducible] (2)

The structural shocks u, are normally distributed, i.e., [u.sub.i]~N(0, [SIGMA]), where [SIGMA] is generally assumed to be a diagonal matrix, usually the identity matrix, such that [u.sub.t] ~ N(0, I). However, equation (2) cannot be estimated directly due to identification issues. Instead, we estimate an unrestricted VAR of the form:

[mathematical expression not reproducible] (3)

[mathematical expression not reproducible] (4)

To recover equation (2) from equation (3), restrictions are imposed on the VAR system to identify an underlying structure. The random stochastic residual [A.sup.-1][Bu.sub.t] is estimated from the residual e, of the estimated VAR. Comparing the residuals from equations (2) and (4), we find that:

[mathematical expression not reproducible] (5)

or equivalently, that:

[mathematical expression not reproducible] (6)

Matrices A and B are invertible of order n. By imposing structure on the matrices A and B, restrictions on the structural VAR in equation (2) are applied. Since E([u.sub.t][u.sub.t]) - [I.sub.n] (the identity matrix) by assumption, equation (5) can be reformulated as:

[mathematical expression not reproducible] (7)

[mathematical expression not reproducible] (8)

Before analysing the effects of the US unconventional monetary policy and its normalization on specific capital inflows, the study proceeds to inspect the stochastic properties of the series considered in the model: first, by calculating the descriptive statistics and plotting the time series of the variables, and second, by selecting suitable unit root tests. The study also performs the requisite Augmented Dickey Fuller (ADF) test. (18)

To determine the appropriate lag length of the individual VAR (p) model, the lag length selection criteria test is conducted. The three criteria (i.e., Akaike Information Criterion (AIC), Schwartz Information Criterion (SIC), and Hannan- Quinn Criterion (HC)) suggest the use of lag length 1. Thus, the VAR (1) models are estimated.

Residual tests (autocorrelation LM and normality Cholesky of Covariance (Lutkephol)) are also generated. To satisfy the stability of the VAR (1) model, the VAR stability condition is checked to ensure that no root lies outside the unit circle. Restrictions are then imposed to identify the VAR (1) system. (19) Furthermore, the impulse response functions (IRFs) of the SVAR system are generated over the twenty four-month horizon by imposing the individual shock to USLIQ (to measure the impact of QE), USSIG (to measure the impact of a surprise monetary policy change relating to normalization), and USFUT (to measure the impact of a change in the future path of the US monetary policy).

For the identification of the AB model, at least [k.sup.2] + k(k-1)/2 = k(3k-1)/2 restrictions are needed, where k denotes the number of variables. If the model is overidentified, which is the case in the empirical application below, the value of a likelihood ratio (LR) is reported. The following assumptions are adopted while imposing structural restrictions to identify the "push" and "pull" factors in the SVAR system:

1. Shocks to other variables in the system have no contemporaneous impact on USLIQ. It is one of the most exogenous variables in the system. This follows the argument by Cushman and Zha (1997) that such restrictions may be acceptable as monetary policy in the US is less likely to reflect foreign shocks.

2. Shocks to other variables in the system have no contemporaneous impact on USSIG. This is another exogenous variable in the system. Meanwhile, USSIG (i.e., taper tantrum, start of tapering, end of QE, and initial Fed rate hike), a surprise component of the US monetary policy normalization, could affect interest rates along the nominal spot and forward yield curves (Cesa-Bianchi, Thwaites and Vicondoa 2016). Hence, it follows that shocks to other variables in the system, except USLIQ and USSIG, have no impact on USFUT.

3. Shocks to other variables in the system have no contemporaneous impact on USVIX, except USLIQ, USSIG and USFUT. The regular release by the US Fed about the future path of the fed funds rate (or USFUT) could influence short-term interest rates. These revised expectations are then factored into investors' investment decisions given their forward-looking behaviour, moving asset prices and affecting USVIX (Doh and Connolly 2013).

4. Shocks to other variables in system, except USLIQ, USSIG, USFUT and USVIX, have no impact on PHSTR. It is assumed that all the global variables have contemporaneous impact on PHSTR. This follows from the argument by Rafiq (2015) that the US unconventional monetary policy and its normalization impact global monetary conditions and subsequently affect global interest rates and risk premia.

5. Shocks to other variables in system, except USLIQ, USSIG, USFUT, USVIX and PHSTR, have no impact on PHRER. This follows that PHSTR affects PHRER contemporaneously (BOE 2016).

6. Shocks to other variables in system, except USLIQ, USSIG, USFUT, USVIX, PHSTR and PHRER, have no impact on PHIPI. This follows the argument that exchange rate influences consumer and business demand, hence, affecting the country's domestic production (or PHIPI) contemporaneously (BOE 2016).

7. Shocks of all the variables in system are assumed to affect [V.sub.i]. Thus, it is assumed to be determined endogenously in the system (Culha 2006; Korap 2010; Lim, Mohapatra, and Stocker 2014).

With ninety-three restrictions in the AB-model, the SVAR is over-identified. Hence, the AB-model used in this study can be expressed as follows:

[mathematical expression not reproducible]

In this study we are interested in examining the effect of shocks from USLIQ, USSIG, and USFUT to VI.

4. Empirical Results and Discussion

4.1 Impact of the Unconventional Monetary Policy and Policy Normalization

This paper finds that the US unconventional monetary policy and its normalization affect portfolio equity, portfolio debt securities, and bank and money market capital flows. Results presented in Figure 7 show that USLIQ has a larger effect, in terms of magnitude of inflows, on portfolio debt securities and bank and money market flows relative to portfolio equity inflows. A positive one-standard deviation shock to USLIQ contracts (20) portfolio debt securities-to-GDP flows by 0.22 per cent in the first month (exhibiting a one period delay in the impact of QE), and increases by 0.24 per cent in the second month. On the other hand, a positive one-standard deviation shock to USLIQ increases bank and money market-to-GDP inflows by 0.23 per cent in the first month, and 0.14 per cent in the second month. Meanwhile, applying the same shock, portfolio equity inflows-to-GDP increases by 0.01 to 0.06 per cent in the first to third month. Among all capital flows, the impact of the shock dies down in the third month.

A Fed rate hike is associated with a reduction in portfolio equity, portfolio debt, and bank and money market capital inflows. Based on the results, a positive one-standard deviation shock to USSIG is estimated to reduce inflows into portfolio equity, portfolio debt securities, and bank and money market, as a percentage of GDP by 0.16 per cent, 0.28 per cent, and 0.54 per cent, respectively, in the second month. In terms of persistence, the impact of the aforementioned shock to portfolio equity dissipates after the first month, and for both portfolio debt securities, and bank and money market flows after the second month (see Figure 7). Meanwhile, a positive one-standard deviation shock to USFUT is estimated to reduce inflows into portfolio equity, portfolio debt securities, and bank and money market, as a percentage of GDP by 0.16 per cent in the first month, 0.25 per cent in the first month and 0.63 per cent in the second month, and 0.16 per cent in the first month and 0.60 per cent in the second month, respectively (see Figure 7). In terms of persistence, the impact of the aforementioned shock to portfolio equity, portfolio debt securities, and bank and money market, dissipates after the first month, third month, and sixth month, respectively.

A monetary policy surprise such as the taper tantrum, start of tapering, end of QE, and first Fed rate hike (as captured by the dummy variable USSIG), has a relatively higher impact on the three specific capital flows in terms of outflows, compared to the impact of a shift in monetary policy rate expectations (i.e., USFUT). Furthermore, results show that, for shocks to USSIG and USFUT, portfolio debt securities, and bank and money market are the most affected flows as shown by the higher magnitude of reversals; meanwhile, portfolio equity flows are the least affected. These results are consistent with the findings of Koepke (2018), indicating that capital flows react more to a surprise change in the US monetary policy and that bond flows react strongly to the said shock compared to equity flows.

Results presented in Table 2 show the generated impulse response functions (IRFs) of each variables of interest to shocks to USLIQ, USSIG and USFUT. A positive sign (+) implies that a specific shock affects the variable of interest positively. A negative sign (-) means that a specific shock affects the variable of interest negatively. The magnitude of shock is the percentage increase or decrease in specific variable of interest (i.e., amount of capital inflows scaled as a percentage of nominal GDP). The persistence of shock to each variable of interest is indicated by the number of months (for instance, M4 means the shock lasts for four months) until specific capital inflow goes back to its stable path.

Among capital flows, the findings in Table 2 (Column 3) show that bank and money market and portfolio debt securities flows face higher risks with QE and policy normalization compared to portfolio equity. This is shown by the contrast between the large magnitude of the response of bank and money market and portfolio debt securities capital flow to impulse from the US policy on the one hand, and the relatively small impact of the US policy on portfolio equity on the other hand. In general, these results are consistent with the findings of Suh and Koo (2016) and Dahlhaus and Vasishtha (2014); both studies observe that portfolio debt, and bank and money market inflows are more vulnerable to the US monetary policy shocks compared to portfolio equity. Similarly, the empirical results also highlight the findings of De Gregorio (2014), who notes the important role of portfolio and banking flows during the GFC, more specifically the severe retrenchment of debt flows.

In terms of magnitude, however, the impact is found to be minimal when the flows are examined as a ratio of GDP. The impulse response functions generated from the SVAR show that the impact of shocks from US Fed policy (i.e., USLIQ, USSIG and USFUT) ranges from 0.03 per cent to 0.69 per cent of capital flows to GDP (see column 2 in Table 2). In addition, the impact does not last very long. The results of the IRF show that the impact of shock from the US policy does not exhibit long periods of persistence. Shocks dissipate at most after the sixth month (see column 3 in Table 2).

Although the empirical findings show that the magnitude of the effect is relatively small and fleeting, capital reversals--particularly those triggered by US policy normalization--can adversely affect an economy if their exposure is large. The results show that sudden reversals brought by US policy normalization is more pronounced than surges brought by QE. In particular, the results of the IRF in Table 2 suggest that the potential impact of US monetary policy normalization (i.e., shocks to USSIG and USFUT), especially a surprise change in monetary policy (i.e., shock to USSIG), is greater than the impact of QE (i.e., shock to USLIQ).

4.2 Variance Decomposition

The impact of US monetary policy on capital flows is further supported by the results of the variance decomposition analysis, which highlight that US policy variables ("push" factors) are the dominant drivers of capital flows as opposed to domestic variables ("pull" factors). Tables 3, 4 and 5 show the variance decomposition of a shock to USLIQ on models with portfolio equity, portfolio debt securities, and bank and money market inflows. Variance decomposition of a shock to USSIG and USFUT on models with portfolio equity, portfolio debt securities, and bank and money market inflows are shown in Appendix B.

When the overall effect of domestic and global factors is considered, it is estimated that the "push" factors jointly account for around 20 per cent of the variations of portfolio equity inflows. Meanwhile, "pull" factors explain around 7 per cent of the forecast error variance of the portfolio equity inflows (Table 3). The results also indicate that, among the "push" and "pull" variables, USVIX (or the US share volatility index) best determines the variance in portfolio equity inflows, which explains nearly 13 per cent. This is followed by USSIG (or the surprise change in monetary policy), which explains around 5 per cent of the variance. The rest of the forecast error variance for models with portfolio equity inflows is accounted for by their own shocks.

Similarly, around 14 per cent of the variance of portfolio debt securities inflows is estimated to be accounted for by "push" factors when the overall effect of domestic and global factors is considered (see Table 4). "Pull" factors explain around 8 per cent of the forecast error variance of the portfolio debt securities inflows. The rest of the forecast error variance for models with portfolio equity inflows is accounted for by their own shocks.
APPENDIX A--cont'd

Variable                   Description

                           Global or "Push " Factors
US Fed funds futures       The US fed funds futures rate at
                           thirty-sixmonth
rate (USFUT)               horizon is used to capture future interest
                           rate expectations. The fed funds futures
                           contract
                           represents the average daily fed funds
                           effective
                           rate for a given calendar month as
                           calculated and
                           reported by the Federal Reserve Bank of New
                           York. It is designed to capture the
                           market's need
                           for an instrument that reflects the Federal
                           Reserve
                           monetary policy.
US Chicago Board           The US Chicago Board Options Exchange
Options Exchange           volatility index is calculated from a
                           weighted
volatility index (USVIX)   average of implied volatilities of various
                           options
                           on the S&P 500 index.
                           Domestic or "Pull" Factors
Philippine short-term      Short-term interest rates are the rates at
                           which
interest rate (PHSTR)      short-term borrowings are affected between
                           financial institutions or the rate at which
                           short-term
                           government paper is issued or traded in the
                           market.
                           Short-term interest rates are generally
                           averages of
                           daily rates, measured as a percentage.
                           Short-term
                           interest rates are based on three-month
                           money
                           market rates. Typical standardized names are
                           "money market rate" and "treasury bill
                           rate". Data
                           are in percentage change.
Philippine real effective  Real effective exchange rate is
                           calculated as
exchange rate (PHRER)      geometric weighted averages of bilateral
                           exchange
                           rates adjusted by relative consumer prices.
                           Changes in the REER thus take into account
                           both
                           nominal exchange rate developments and the
                           inflation differential vis-a-vis trading
                           partners.
Philippine industrial      Industrial production index is used as
                           proxy for
production index (PHIPI)   GDP. The industrial production index
                           measures
                           the change in output in major manufacturing
                           industries.

Variable                   Source

US Fed funds futures       Bloomberg (2016)
rate (USFUT)
US Chicago Board           Federal Reserve
Options Exchange           Bank of St. Louis
volatility index (USVIX)   Economic Research
                           (FRED) (2016)
Philippine short-term      BSP (2016)
interest rate (PHSTR)
Philippine real effective  BSP(2016)
exchange rate (PHRER)
Philippine industrial      BSP (2016)
production index (PHIPI)


APPENDIX B Variance Decomposition

B.1: Shocks to USSIG
TABLE 6
Variance Decomposition of Model with Portfolio Equity Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PH1PI

1                0.07   4.51   1.92   13.08  1.04   3.12   1.64
6                0.20   4.51   1.99   12.88  1.11   4.79   1.67
12               0.21   4.51   1.99   12.87  1.17   4.78   1.67
18               0.22   4.50   1.99   12.87  1.20   4.78   1.67
24               0.23   4.50   1.99   12.86  1.22   4.78   1.67

Variance Period  SPHPE

1                74.61
6                72.86
12               72.81
18               72.77
24               72.75

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PH1PI.
Source: Authors' calculations.

TABLE 7
Variance Decomposition of Model with Portfolio Debt Securities Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIPI

 1               0.34   2.15   0.87   0.23   1.89   4.78   1.08
 6               0.78   5.34   3.90   4.08   1.95   4.53   1.43
12               0.81   5.34   3.90   4.08   2.04   4.53   1.43
18               0.82   5.34   3.90   4.07   2.10   4.52   1.43
24               0.83   5.33   3.90   4.07   2.13   4.52   1.42

Variance Period  SPHPD

 1               88.64
 6               77.99
12               77.88
18               77.82
24               77.79

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PHIPI.
Source: Authors' calculations.

TABLE 8
Variance Decomposition of Model with Bank and Money Market Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIPI

 1               0.53   0.55   0.23   3.36   3.51   1.33   0.01
 6               0.72   1.47   3.32   6.00   3.29   1.89   0.15
12               0.73   1.47   3.32   6.00   3.29   1.89   0.15
18               0.73   1.47   3.32   6.00   3.29   1.89   0.15
24               0.73   1.47   3.32   6.00   3.29   1.89   0.15

Variance Period  SPHBM

 1               90.47
 6               83.16
12               83.16
18               83.15
24               83.15

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PHIPI.
Source: Authors' calculations.


B.2: Shocks to USFUT
TABLE 9
Variance Decomposition of Model with Portfolio Equity Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIP1

 1               0.07   4.51   1.92   13.08  1.04   3.12   1.64
 6               0.20   4.51   1.99   12.88  1.11   4.79   1.67
12               0.21   4.51   1.99   12.87  1.17   4.78   1.67
18               0.22   4.50   1.99   12.87  1.20   4.78   1.67
24               0.23   4.50   1.99   12.86  1.22   4.78   1.67

Variance Period  SPHPE

 1               74.61
 6               72.86
12               72.81
18               72.77
24               72.75

Notes: Push factors: USLIQ. USSIG, USFUT, USVIX; Pull factors: PHSTR.
PHRER. PHIPI.
Source: Authors' calculations.

TABLE 10
Variance Decomposition of Model with Portfolio Debt Securities Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIPI

 1               0.34   2.15   0.87   0.23   1.89   4.78   1.08
 6               0.78   5.34   3.90   4.08   1.95   4.53   1.43
12               0.81   5.34   3.90   4.08   2.04   4.53   1.43
18               0.82   5.34   3.90   4.07   2.10   4.52   1.43
24               0.83   5.33   3.90   4.07   2.13   4.52   1.42

Variance Period  SPHPD

 1               88.64
 6               77.99
12               77.88
18               77.82
24               77.79

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PHIPI.
Source: Authors' calculations.

TABLE 11
Variance Decomposition of Model with Bank and Money Market Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIPI

 1               0.30   0.54   0.22   3.17   3.54   1.29   0.03
 6               0.64   1.46   3.31   5.65   3.32   1.86   0.17
12               0.65   1.46   3.31   5.65   3.32   1.86   0.17
18               0.65   1.46   3.31   5.65   3.33   1.86   0.17
24               0.65   1.46   3.31   5.65   3.33   1.86   0.17

Variance Period  SPHBM

 1               90.91
 6               83.59
12               83.58
18               83.58
24               83.58

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PHIPI.
Source: Authors' calculations.


NOTES

Views, findings, and conclusions expressed in this study are entirely of the authors and should not be attributed to the Bangko Sentral ng Pilipinas and the International Monetary Fund, its Management, or its Executive Directors.

(1.) "Push" factors are global determinants affecting capital into EMEs (e.g., common/global shocks to risk, liquidity, asset prices, among others) while "pull" factors are country-specific drivers of capital inflows into EMEs (e.g., country-specific shocks, institutional quality, country risk rating, and macroeconomic fundamentals) (Fratzcher 2011).

(2.) The ASEAN-5 is composed of Indonesia, Malaysia, the Philippines, Singapore and Thailand.

(3.) Source: BSP (2016); and Butocan (2009).

(4.) Data on gross private capital inflows are derived from the balance of payments (BOP). The BOP is comprised of two main groups of accounts: the current account, and the capital and financial account. Private capital flows are derived from three standard components of the financial account: direct investments, portfolio investments (composed of portfolio equity and portfolio debt securities), and other investments (composed mainly of bank and money market flows). Another component known as financial derivatives is not included as reported figures are largely negligible. Each of the three components separates the transactions between residents and non-residents. For direct investment, there are directional distinctions (abroad or in the reporting country). The portfolio investment and other investment components are divided into assets and liabilities. On this basis, gross private inflows include items recorded as direct investment, its portfolio investment, and other investment liabilities into the Philippines.

(5.) The slow growth was evident in the following advanced economies in certain periods: the US grew by -0.3 per cent in 2008 and -2.8 per cent in 2009; the United Kingdom experienced recession recorded at -0.4 per cent in 2008 and -4.2 per cent in 2009; Japan grew by -1.0 per cent in 2008 and -5.5 per cent in 2009; and Europe grew by 0.7 per cent in 2008 and -4.3 per cent in 2009.

(6.) See BSP (2016) for details about the reforms implemented during the period.

(7.) Ibid.

(8.) Ibid.

(9.) Ibid.

(10.) The BSP prepaid its outstanding obligations to the International Monetary Fund in December 2006 amounting US$220 million, ending its post-programme monitoring (PPM) arrangement with the Fund. Other prepayment measures were also pursued. Similarly, the prepayment of the Philippine Brady Bonds, in April 2007, cleared collateral amounting to US$103 million including interest earned, of which, US$85 million was credited to the BSP and the balance to the National Government.

(11.) See BSP (2016) for specific reforms implemented.

(12.) Ibid.

(13.) Issued under BSP Circular No. 855 Series of 2014.

(14.) The FSSC is a forum composed of the BSP, Department of Finance, Securities and Exchange Commission, and Insurance Commission.

(15.) Similar variables, except USS1G, are also used in previous studies such as Culha (2006), Korap (2010), Dahlhaus and Vasishtha (2014), Lim et al. (2014), Koepke (2015), and Suh and Koo (2016).

(16.) The same variables are also used in previous studies such as Culha (2006), Korap (2010), Lim, Mohapatra and Stocker (2014), and Suh and Koo (2016).

(17.) This time period is selected for the analysis to cover the year when QE measure was first implemented until the period of the latest available data. It is important to note that no significant change is observed in the results when the time period considered starts in November 2008 (i.e., the actual implementation of QE1). First log-difference transformations are applied to all variables, except USSIG, PHSTR, SPHPE, SPHPD and SPHBM.

(18.) Adopted from DF test, the ADF test tests the null hypothesis that a time series y, is 1(1) against the alternative that it is 1(0), assuming that the dynamics in the data have an ARMA structure.

(19.) The SVAR systems are overidentified with 1 degree of freedom. The likelihood ratio (LR) test estimated for the system identification restrictions under the null hypothesis for models with SPHPE; SPHPD; and SPHBM; are Chi-squared(l) = 0.0039 with probability value of 0.9502; Chi-squared(l) = 0.0414 with probability value of 0.8388; and Chi-squared(l) = 0.0199 with probability value of 0.8878; respectively.

(20.) The negative relationship of QE and portfolio debt securities in the first month is consistent with findings of Suh and Koo (2016) indicating that factors other than QE may have affected the said inflows in the opposite direction.

(21.) The USIPI is an economic indicator published by the US Federal Reserve Board that measures the real production output of manufacturing, mining, and utilities, usually used as proxy for GDP. First log-difference transformation is also applied to USIPI.

REFERENCES

Amisano, G. and C. Giannini. 1997, Topics in Structural VAR Econometrics. Heidelberg: Springer-Verlag Berlin. BSR 2016. BSP Annual Report 1999-2015. http://www.bsp.gov.ph/publications/regular asp (accessed 15 August 2016).

BOE. 2016. How Does Monetary Policy Work? http://www.bankofengland.co.uk/monetarypolicy/Pages/how.aspx (accessed 25 June 2016).

Bauer, M. and G. Rudebusch. 2013. Expectations for Monetary Policy Liftoff. FRBSF Economic Letter 2013-34. Federal Reserve Bank of San Francisco.

Butocan. S. 2009. "Capital Flows and Their Implications for Central Bank Policies in the Philippines". In Capital Flows and Implication for Central Bank Policies in the SEACEN Countries, edited by M. Shrestha and C. Lim, pp. 249-86. Kuala Lumpur: The South East Asian Central Banks (SEACEN) Research and Training Centre.

Cesa-Bianchi, A., G. Thwaites, and A. Vicondoa. 2016. "Monetary Policy Transmission in an Open Economy: New Data and Evidence of the United Kingdom". Paper No. CFM-DP2016-12. Centre for Macroeconomics.

Chen, Q., A. Filardo, D. He, and F. Zhu. 2013. "International Spillovers of Central Bank Balance Sheet Policies". BIS Paper 66. Bank for Internationa! Settlements.

Culha, A. 2006. "A Structural VAR Analysis of the Determinants of Capital Flows into Turkey". Central Bank Review. Central Bank of Turkey.

Cushman. D., and T Zha. 1997. "Identifying Monetary Policy in a Small Open Economy Under Flexible Exchange Rates". Journal of Monetary Economics 39: 433-48.

Dahlhaus, T. and G. Vasishtha. 2014. "The Impact of US Monetary Policy Normalisation on Capital Flows to Emerging Economies". Working Paper 2014-53. Banque Du Canada.

De Gregorio, J. 2014. "Capital Flows and Capital Account Management". In What Have We Learned? Macroeconomic Policy After the Crisis, edited by G. Akerlof, O. Blanchard, D. Romer, and J. Stiglitz, pp. 271-85. Cambridge: The MIT Press.

Doh, T. and M. Connolly. 2013. "Has the Effect of Monetary Policy Announcement on Asset Price Changed?". Economic Review, 3rd Quarter 2013. Federal Reserve Bank of Kansas City.

Elliot, G., T Rothenberg, and J. Stock. 1996. "Efficient Tests for an Autoregressive Unit Root". Econometrica 64, no. 4: 813-36.

Fernandez, J. and A. Kalaw. 2016. "Living with Capital Flow Volatility in the Philippines". In Living with Volatilities: Capital Flows and Policy Implications for SEACEN Central Banks, edited by C. Becker, pp. 203-33. Kuala Lumpur: South East Asian Central Banks (SEACEN) Research and Training Centre.

Fratzscher, M. 2011. "Push Factors versus Pull Factors and the Global Financial Crisis". Working Paper Series 1364. European Central Bank.

Fratzscher, M., M. Lo Duca, and R. Straub. 201). "On the International Spillovers of US Quantitative Easing". Working Paper Series 1557. European Central Bank.

Gottschalk, J. 2001. "An Introduction into the SVAR Methodology: Identification, Interpretation and Limitations of SVAR Models". Kiel Working Paper No. 1072. Kiel Institute of World Economics.

Koepke, R. 2018. "Fed Policy Expectations and Portfolio Flows to Emerging Markets". Journal of International Markets, Institutions and Money 55 (2018): 170-94.

Korap, L. 2010. "Identification of 'Pull" & 'Push' Factors for the Portfolio Flows: SVAR Evidence from Turkish Economy". Dogus Universitesi Dergisi 11, no. 2: 223-32. Istanbul University, Institute of Social Science.

Lim, J., S. Mohapatra, and M. Stocker. 2014. "Tinker, Taper, QE, Bye? The Effect of Quantitative Easing on Financial Flows to Developing Countries". Policy Research Working Paper Series 6820. World Bank.

Rafiq, S. 2015. "The Effects of Unconventional Monetary Policy on Asia Frontier Developing Economies". IMF Working Paper WP/15/18. International Monetary Fund.

Rummel, O. 2015. "Economic Modelling and Forecasting: Subset VAR. SVAR, and VECM Modelling". Centre for Central Banking Studies, Bank of England.

Suh, S. and B. Koo. 2016. "Spillovers from US Unconventional Monetary Policy and Its Normalization to Emerging Markets: A Capital Flow Perspective". BOK Working Paper No. 2016-4. Bank of Korea.

Sun. T. 2015. "The Impact of Global Liquidity on Financial Landscapes and Risks in the ASEAN-5 Countries". IMF Working Paper WP/15/211. International Monetary Fund.

Ivy G. Sabuga and Jacinta Bernadette Rico Shirakawa

Ivy G. Sabuga is Bank Officer V, Department of Economic Research, Bangko Sentral ng Pilipinas, A. Mabini St. cor. P. Ocampo St.. Malate Manila. Philippines 1004; and PhD Economics candidate, City, University of London, Northampton Square, Clerkenwell, London EC1V, United Kingdom; email: sabugaig@bsp.gov.ph; ivy.sabuga@city.ac.uk

Jacinta Bernadette Rico Shirakawa is Economist at the International Monetary Fund, 1900 Pennsylvania Ave NW. Washington, DC, 20431, USA; email: jshirakawa@imf.org. This research was conducted while the author was with Waseda University, Japan and before she joined the International Monetary Fund (IMF).

The views expressed in this paper are those of the author(s) and do not necessarily represent the views of the Bangko Sentral ng Pilipinas, or the IMF, its Executive Board, or IMF management.

DOI: 10.1355/ae37-2b
TABLE 1
Magnitude of Capital Inflows into the Philippines (in US$ million,
1999-2015)

Items                          1999      2000    2001         2002

Gross private capital inflows  5,801     1,663         20     1,999
Percentage of GDP                  6.99      2.05       0.03      2.46
Foreign direct investments     1,247     2,240        195     1,542
Percentage of GDP                  1.50      2.76       0.26      1.90
Portfolio equity                 489      -202        125       227
Percentage of GDP                  0.59     -0.25       0.16      0.28
Portfolio debt securities      3,429       461        959     1,147
Percentage of GDP                  4.13      0.57       1.26      1.41
Bank and money market            636      -836     -1,259      -917
Percentage of GDP                  0.77     -1.03      -1.65     -1.13

Items                          2003      2004       2005      2006

Gross private capital inflows  1,114       -131     4,691      4,028
Percentage of GDP                  1.33      -0.14      4.55       3.30
Foreign direct investments       491        688     1,664      2,707
Percentage of GDP                  0.59       0.75      1.61       2.22
Portfolio equity                 500        518       420      1,348
Percentage of GDP                  0.60       0.57      0.41       1.10
Portfolio debt securities        880     -1,321     2,521      2,231
Percentage of GDP                  1.05      -1.45      2.45       1.83
Bank and money market           -757        -16        85     -2,258
Percentage of GDP                 -0.90      -0.02      0.08      -1.85

Items                          2007      2008        2009

Gross private capital inflows  9,188     -5,314      3,887
Percentage of GDP                  6.15      -3.06       2.31
Foreign direct investments     2,919      1,340      2,065
Percentage of GDP                  1.95       0.77       1.23
Portfolio equity               1,137       -462        308
Percentage of GDP                  0.76      -0.27       0.18
Portfolio debt securities        580-     -2,730     1,980
Percentage of GDP                  0.39       -1.57      1.18
Bank and money market          4,552      -3,462      -466
Percentage of GDP                  3.05       -1.99      -0.28

Items                          2010       2011      2012       2013

Gross private capital inflows  13,058     6,449     10,857     4,331
Percentage of GDP                   6.54      2.88       4.34      1.59
Foreign direct investments      1,070     2,007      3,215     3,737
Percentage of GDP                   0.54      0.90       1.29      1.37
Portfolio equity                  833     1,040      1,753       -34
Percentage of GDP                   0.42      0.46       0.70     -0.01
Portfolio debt securities       5,526     2,060      2,417       397
Percentage of GDP                   2.77      0.92       0.97      0.15
Bank and money market           5,629     1,341      3,473       230
Percentage of GDP                   2.82      0.60       1.39      0.08

Items                           2014      2015

Gross private capital inflows   5,670      4,585
Percentage of GDP                   1.99       1.57
Foreign direct investments      5,740      5,724
Percentage of GDP                   2.02       1.96
Portfolio equity                1,196       -756
Percentage of GDP                   0.42      -0.26
Portfolio debt securities      -1,199     -1,385
Percentage of GDP                  -0.42      -0.47
Bank and money market             -66      1,002
Percentage of GDP                  -0.02       0.34

Notes: Based on authors' calculation using the Balance of Payments
(BOP) data. Data from 1999 to 2004 are compiled using the Balance of
Payments Manual Five (BPM5) framework while data from 2005 to present
are compiled using the Balance of Payments Manual Six (BPM6) framework.
The gross domestic product (GDP) used to scale the capital inflows is
based on current prices.

TABLE 2
Summary of Results on the Impact of the US Unconventional Monetary
Policy and Its Normalization on Capital Inflows into the Philippines

Variable of Interest  Magnitude (in per cent)
                      Shock to USLIQ

SPHPE                  0.01(M1); 0.01 (M2); 0.06(M3)
SPHPD                 -0.22(M1); 0.24(M2); -0.34(M3)
SPHBM                  0.23(M1); 0.14(M2); 0.34(M3)
                      Shock to USSIG
SPHPE                 -0.03(M1); 0.16(M2)
SPHPD                 -0.26(M1);-0.28(M2)
SPHBM                 -0.22(M1);-0.54(M2)
                      Shock to USFUT
SPHPE                 -0.16(M1)
SPHPD                 -0.25(M1); -0.63(M2); -0.13(M2)
SPHBM                 -0.16(M1); -0.60(M2); -0.34(M3); -0.14(M4);
                      -0.06(M5); -0.03(M6)

Variable of Interest  Persistence

SPHPE                 M3
SPHPD                 M3
SPHBM                 M3
SPHPE                 M2
SPHPD                 M2
SPHBM                 M2
SPHPE                 Ml
SPHPD                 M3
SPHBM                 M6

Notes: The confidence intervals estimated for some variables are
large: hence, it is necessary to consider some margins of uncertainty
in interpreting the results. Once the plus and minus one standard
error bands are added, we can see how significant these effects are.
The results show that only the impact of a shock to USLIQ on SPHBM in
period three; USSIG on SPHPE in period two and SPHBM in period three;
and USF UT on SPHPE in period one, SPHPD in period two, and SPHBM
beginning period two (i.e., M2) are significant at +/- 1 S.E.

TABLE 3
Variance Decomposition of Model with Portfolio Equity Inflows

Variance Period  USLIQ  USSIG  USFUT  USVIX  PHSTR  PHRER  PHIP1

1                0.07   4.51   1.92   13.08  1.04   3.12   1.64
6                0.20   4.51   1.99   12.88  1.11   4.79   1.67
12               0.21   4.51   1.99   12.87  1.17   4.78   1.67
IS               0.22   4.50   1.99   12.87  1.20   4.78   1.67
24               0.23   4.50   1.99   12.86  1.22   4.78   1.67

Variance Period  SPHPE

1                74.61
6                72.86
12               72.81
IS               72.77
24               72.75

Notes: Push factors: USLIQ, USSIG, USFUT, USVIX; Pull factors: PHSTR,
PHRER, PHIPI.
Source: Authors' calculations.
COPYRIGHT 2020 Institute of Southeast Asian Studies (ISEAS)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2020 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sabuga, Ivy G.; Shirakawa, Jacinta Bernadette Rico
Publication:Journal of Southeast Asian Economies
Geographic Code:9PHIL
Date:Aug 1, 2020
Words:9641
Previous Article:Indonesian Macro Policy Through Two Crises.
Next Article:Women in Business and Management in Thailand: Transforming High Participation Without EEO.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |