## 11. Evolution of Global Aggregates and Capital Flows, 1996–2011

## Source details

**Canonical URL:** [11. Evolution of Global Aggregates and Capital Flows, 1996–2011](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1409.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1409.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1409.pdf.json)

---

### Introduction
- Purpose: examine relationship between bank liability aggregates and global financial conditions, focusing on bank liabilities as counterparts to banking sector assets and as indicators of credit availability and financial vulnerability.
- Key hypotheses:
  - Rapid credit growth is an indicator of potential financial vulnerability; normalized measures such as credit to GDP ratio matter.
  - The money aggregate associated with claims of non‑financial corporations (NFCs) on the banking sector (denoted L) is closely correlated with financial conditions facing firms operating across borders and with global economic activity.
- Rationale for liability‑side measures:
  - Bank liabilities are more transparent and homogenous than assets; largely short term and close to market values.
  - Liabilities can be organized into core and non‑core with contrasting cyclical properties; non‑core liabilities are more procyclical.
  - NFC activity that straddles borders can create currency mismatches and carry‑trade like positions not captured well by residence‑based external debt statistics.
- Data source: standardized reporting forms (SRFs) submitted to the IMF and published in International Financial Statistics (IFS).

### Data definition, construction, and caveats
- Definitions:
  - L: sum of deposits of NFCs in the banking system (“other depository corporations (ODCs)”), including transferable and other deposits included in or excluded from broad money definitions.
  - GL (Global Liquidity): sum across countries of each country’s L converted into a common numeraire currency (e.g., U.S. dollar GL denoted in the source).
- Data limitations:
  - SRFs do not provide consistent MMF or shadow‑banking claims; L includes only deposits in the regulated banking sector.
  - Quality of monetary statistics in SRFs is uneven; some series show discontinuities and unexplained jumps.
  - Some key countries do not submit full SRF monetary data.
- Numeraire choice matters: the U.S. dollar GL measure incorporates exchange rate movements of the U.S. dollar vis‑à‑vis other currencies; numeraire choice affects information content of GL.

### Descriptive patterns and key statistics
- Time series patterns and stylized facts:
  - Until around 2004 GL did not exceed 15 percent of global broad money; since then NFC deposits became a larger proportion of global broad money, with a sharp fall during the 2008 crisis.
  - U.S. dollar GL and Japanese yen GL series are more similar; Euro GL can differ substantially.
  - U.S. dollar GL for middle income countries is more procyclical than full sample; U.S. dollar GL for middle income countries declines close to 30 percent in 2009Q1 year‑on‑year.
  - NFC deposit growth measured in domestic currency remains positive at the crisis trough.
- Selected sample and frequency:
  - Panel regressions sample period: 2001Q4–2013Q1; quarterly; up to 88 countries in some regressions.
- Summary statistics (selected exact values from Table 8):
  - VIX (Quarter): Obs = 70; Mean = 3.0161; Std. Dev. = 0.3371; Min = 2.4006; Max = 4.0707.
  - ΔVIX (Quarter): Obs = 69; Mean = 0.0074; Std. Dev. = 0.2134; Min = -0.4978; Max = 0.8491.
  - Δ Interoffice (Quarter): Obs = 65; Mean = -0.0001; Std. Dev. = 1.1145; Min = -4.1770; Max = 5.2829.
  - ΔM0 (Quarter): Obs = 3,011; Mean = 0.0269; Std. Dev. = 0.1602; Min = -3.1091; Max = 2.9787.
  - ΔM2 (Quarter): Obs = 2,977; Mean = 0.0308; Std. Dev. = 0.1320; Min = -2.3519; Max = 2.6670.
  - Capital Inflow (Quarter): Obs = 2,970; Mean = 0.0235; Std. Dev. = 0.1100; Min = -0.7631; Max = 0.9285.
  - Real GDP Growth (Annual): Obs = 810; Mean = 0.0326; Std. Dev. = 0.0376; Min = -0.1480; Max = 0.1420.
  - Inflation (Annual): Obs = 802; Mean = 0.0679; Std. Dev. = 0.2507; Min = -0.1285; Max = 5.4921.
  - Δ Debt to GDP (Annual): Obs = 700; Mean = 0.0043; Std. Dev. = 0.0773; Min = -0.3991; Max = 1.1137.

### Global liquidity as an activity indicator — empirical panel results (summary)
- Sample and regressions:
  - Quarterly panel regressions for up to 88 countries; dependent variables: country GDP growth (Table 1), imports (Table 2), exports (Table 3).
- Main empirical findings (preserve coefficients and significance as reported):
  - Lagged Dlog(Global NFC in USD) (t-1) associated with real GDP growth: 1,093*** [1.39e-07]; 1,138*** [4.89e-09] (Table 1 examples).
  - Dlog(Global NFC in JPY) (t-1): 1,021*** [7.84e-09]; 1,011*** [1.95e-08] (Table 1 examples).
  - Dlog(Broad Money) coefficients (examples): 0.0521** [0.0144]; 0.0562** [0.0171]; 0.0599*** [0.00705]; 0.0661*** [0.00677]; 0.0563*** [0.00876]; 0.0612*** [0.00957].
  - QE dummy on GDP growth (examples): -0.0277*** [2.58e-09]; -0.0282*** [2.14e-10]; -0.0249*** [7.83e-09]; -0.0253*** [8.26e-10]; -0.0216*** [1.44e-07]; -0.0220*** [1.17e-08].
  - For imports (Table 2): Dlog(Global NFC in USD) (t-1) examples: 8,230*** [0]; 6,757*** [0]; Dlog(Global NFC in JPY) (t-1): 5,266*** [0]; 4,482*** [0].
  - For exports (Table 3): Dlog(Global NFC in USD) (t-1): 6,296*** [0]; 4,887*** [5.51e-08]; Dlog(Global NFC in JPY) (t-1): 4,406*** [5.63e-10]; 3,279*** [5.47e-06].
  - Individual country NFC deposit growth (Dlog(NFC) (t-1)) often enters negative for imports/exports: examples include -0.0461** [0.0122]; -0.0478*** [0.00994]; -0.0542** [0.0165]; -0.0513** [0.0241].
  - Growth of U.S. broker dealer sector (proxy for global bank credit supply) enters with a positive and highly significant coefficient across regressions (as noted in text summary).
  - QE*VIX interactions: QE coefficient negative and significant for GDP growth; QE*VIX has positive coefficient for GDP growth (mitigating growth when VIX is high) and often negative for imports/exports.
- Interpretation:
  - U.S. dollar (and JPY) GL capture funding currency dynamics that amplify quantity contractions through exchange rate movements; choice of numeraire matters.
  - Global GL acts as a common global factor explaining co‑movement of activity across countries.

### Determinants of NFC deposits — country‑level regressions (summary)
- Objective: explain quarterly growth of individual country NFC deposits (in domestic currency) across income group subsamples (Tables 4–7).
- Key explanatory variable: Capital Inflows = growth in financial liabilities of “other sectors” in BOP (debt security liabilities and loan liabilities of NFCs and OFCs).
- Full sample (Table 4) summary:
  - Capital flows variable marginally significant contemporaneously and with four-quarter lag.
  - Exchange rate change sometimes negative (appreciating domestic currency associated with higher NFC deposit growth).
  - VIX enters with negative sign: higher volatility associated with slower NFC deposit growth.
- Middle income subsample (Appendix 2/Table 5) key exact reported values:
  - Capital inflows (t-4): 0.0108*** [0.00328]; 0.0108*** [0.00311]; 0.0102*** [0.00621] across columns.
  - ∆ Nominal Exchange rate (t-1): -2.03e-05* [0.0599]; -2.08e-05** [0.0298]; -2.71e-05* [0.0762].
  - Log(VIX) (t-1): -0.0214*** [0.00604]; -0.0200** [0.0126]; -0.0201** [0.0190].
  - Dlog(VIX) (t-1): 0.0233* [0.0949]; 0.0291* [0.0554]; 0.0227 [0.101].
  - Growth of interoffice (t-1): 0.0120* [0.0509] (one specification).
  - QE * VIX (t-1): -20.84* [0.0932] (one specification); QE * VIX (CIS): -18.19*** [0.00581]; QE * VIX (SSA): 9.139*** [0.00687].
  - Observations reported: 535 (and alternative columns with 1,704); R-squared: 0.029; 0.034; 0.044; Number of country_code: 38; 48.
- High income subsample (Table 6) key exact values:
  - Dlog(BIS external loans nonbanks) (t-1): 0.0659* [0.0694] (one specification).
  - Dlog(BIS external loans nonbanks) (t-4): 0.125** [0.0200]; 0.108** [0.0370]; 0.104* [0.0520].
  - QE(t-1): -0.0746* [0.0643] (one specification).
  - Inflation (t-1): 0.0116* [0.0898] (one specification).
  - Observations: 124 and 309 in columns; R-squared vary (examples: 0.042; 0.067; 0.056; 0.035; 0.042; 0.043); Number of country_code: 9.
- Low income subsample (Table 7) key exact values:
  - Dlog(VIX) (t-1): 0.0987*** [0.00593] (one specification).
  - QE(t-1): 0.0554* [0.0688] in one specification.
  - Δ Debt/GDP (t-1): -9.961* [0.0659]; -13.32*** [0.00861]; -13.79** [0.0147] (negative and significant in fixed-effect specifications).
  - Observations: 353; 472; R-squared: 0.172; 0.199; 0.196 (fixed-effect columns) and 0.004; 0.004; 0.005 (other columns); Number of country_code: 7; 14.

### Determinants of M0 and M2 growth (Tables 9a/9b and 10a/10b)
- General panel setup and variable definitions (preserved exactly):
  - Dependent variables: quarterly log difference (growth) of money stock—either M2 or M0.
  - Specification examples:
    - ∆M_{c,t} = β0 + β1 Capital_Inflow_{c,t-1} + β2 ∆RER_{c,t-1} + β3 VIX_{t-1} + β4 ∆VIX_{t} + β5 ∆Interoffice_{t-1} + control_{c,t} + e_{c,t}
    - ∆M_{c,t} = β0 + β1 Capital_Inflow_{c,t-1} + β2 ∆RER_{c,t-1} + β3 VIX_{t-1} + β4 ∆VIX_{t} + β5 ∆Interoffice_{t-1} + β6 QE_dummy + control_{c,t} + e_{c,t}
  - QE_dummy: QE 1: 2009 Q1-2010 Q1; QE 2: 2010 Q4-2011 Q2.
- M0 results highlights (Tables 9a/9b, preserved exact reported coefficients and p‑values where cited):
  - M0 Growth(t-1): -0.2986*** [0.0650]; -0.2987*** [0.0649]; -0.2143*** [0.0808]; -0.2160*** [0.0807] (selected columns).
  - Capital Inflow(t-1): 0.0251** [0.0117]; 0.0252* [0.0117]; 0.0269** [0.0127]; 0.0266** [0.0127]; 0.0272* [0.0148]; 0.0273* [0.0149]; 0.0294* [0.0164]; 0.0288* [0.0164].
  - VIX(t-1): 0.0144*** [0.0035]; 0.0158*** [0.0036]; in other columns 0.0011 [0.0046]; 0.0015 [0.0045].
  - △Interoffice(t-1): 0.0075*** [0.0022]; 0.0075*** [0.0022]; 0.0054** [0.0022]; 0.0051** [0.0021]; 0.0082*** [0.0022]; 0.0082*** [0.0022]; 0.0078*** [0.0021]; 0.0072*** [0.0019].
  - QE: -0.0159* [0.0086] (one System GMM specification).
  - Real GDP Growth(t-1): 0.1186** [0.0510]; 0.1118* [0.0568]; 0.2696*** [0.0693]; 0.2281*** [0.0792].
  - Inflation(t-1): 0.1025*** [0.0359]; 0.1010*** [0.0357]; 0.1995*** [0.0438]; 0.1876*** [0.0403].
  - Observations and countries (selected): Observations: 2,646; 2,462; 2,134 across columns; # of Countries: 45.
  - AR(1) p-values: 0.0001; 0.0001; 0.0006; 0.0006. AR(2) p-values: 0.0603; 0.0670; 0.9818; 0.9673.
- M2 results highlights (Tables 10a/10b, preserved exact reported coefficients and p‑values where cited):
  - M2 Growth(t-1): -0.3838*** [0.0800]; -0.3831*** [0.0799]; -0.2997*** [0.1151]; -0.2995*** [0.1140] and similar in Table 10b.
  - Capital Inflow(t-1): 0.0145* [0.0082]; 0.0146* [0.0082]; 0.0178* [0.0101]; 0.0177* [0.0100]; 0.0199* [0.0106]; 0.0200* [0.0106]; 0.0336** [0.0164]; 0.0338** [0.0164] (Table 10a). Table 10b includes 0.0349* [0.0199]; 0.0360* [0.0202] in some columns.
  - △RER(t-1): -0.0420* [0.0245]; -0.0431* [0.0250]; -0.0312* [0.0170]; -0.0321* [0.0172] (selected columns).
  - VIX(t-1): mixed signs with significance examples: -0.0072** [0.0030]; -0.0064* [0.0030]; 0.0151** [0.0029]; 0.0163*** [0.0031].
  - QE: -0.0127*** [0.0045] (one specification); other columns show QE coefficients like -0.0037 [0.0025]; -0.0095** [0.0037].
  - Real GDP Growth(t-1): 0.1642*** [0.0262]; 0.1518*** [0.0288]; 0.4003*** [0.0603]; 0.3648*** [0.0579].
  - Inflation(t-1): 0.0805*** [0.0181]; 0.0778*** [0.0186]; 0.1397*** [0.0356]; 0.1296*** [0.0347].
  - Observations and countries (selected): Observations: 2,623; 2,326; 2,134 across columns; # of Countries: 45.
  - AR(1) p-values: 0.0067; 0.0067; 0.0124; 0.0126. AR(2) p-values: 0.2079; 0.2044; 0.6865; 0.6667.

### Time series patterns and graphical evidence (Figure 11 summary)
- Observed dynamics 1996–2011:
  - M2 is procyclical; M0 displays some countercyclical features, especially around the financial crisis.
  - U.S. dollar REER weakening in the mid‑2000s coincided with an uptick in capital inflows as measured by BIS 7B.
  - Global aggregates plotted include Net Interoffice Assets (right axis), Global Capital Inflows (left axis), REER in the United States, Global M0 and M2 growth rates, Global trade and global GDP growth.
  - Figure legend note: "All growth rates are annualized."

### Conclusions and policy implications
- Key conclusions:
  - Specialized bank liability aggregates (e.g., NFC deposits) can have information value as indicators of economic activity because of their sensitivity to global credit conditions.
  - Monetary aggregates, when adapted, can play an indicative role in:
    - Providing insights into credit condition patterns through liability-side aggregates of bank balance sheets.
    - Offering a systemic perspective on financial stability tied to banking sector procyclicality and vulnerability to reversals of capital flows.
- Core versus non‑core liabilities distinction:
  - Core liabilities: funding drawn on during normal times and sourced mainly domestically (retail deposits of the household sector are a useful first conjecture).
  - Non‑core liabilities: volatile funding sources tapped during lending booms (includes short-term foreign currency‑denominated bank liabilities and wholesale corporate deposits).
  - The ratio of non‑core to core liabilities is a reliable indicator of vulnerability to crises (reference in source to Hahm, Shin, and Shin (2013)).
- Recommended adaptations for monitoring and macroprudential policy (preserved framing from source):
  - For countries with open capital markets:
    - International capital flows into the banking sector are key indicators of financial vulnerability.
    - Short-term foreign currency‑denominated bank liabilities can be treated as volatile non‑core liabilities of the banking sector.
  - For countries with relatively closed financial systems:
    - Adapt conventional monetary aggregates by distinguishing who holds the claims: household retail deposits versus corporate deposits.
    - The distinction between retail deposits and NFC wholesale deposits is particularly important.
  - More generally:
    - Use the accounting principle that defines core versus non‑core liabilities to guide classification of financial systems.
    - The relative size of non‑core liabilities can serve as a monitoring tool reflecting the stage of the financial cycle, vulnerability to setbacks, and as an early warning indicator.
- Practical caveats and research implications:
  - Classification into core and non‑core is not clear‑cut (e.g., deposits of small owner‑managed enterprises may resemble household deposits; large firms’ deposits may be wholesale).
  - Traditional monetary aggregates defined by legal form and transactional liquidity may be less effective as macroprudential tools without adaptation to claim‑holder composition.
  - Dynamic panel specifications are appropriate to study causality given quarterly data and persistence of shocks; suggested as follow‑up research.

*Source: IMF Working Paper content from _wp1409 — chapter "11. Evolution of Global Aggregates and Capital Flows, 1996–2011" and Appendices (tables, figures, and regression outputs as provided).*

### References .............................................................................................................

### _wp1409 - References

### Tables
- 1. Global Money Stock as Activity Indicator for real GDP growth ........................................37
- 2. Global Money Stock as Activity Indicator of Imports .........................................................38
- 3. Global Money Stock as Activity Indicator of Export ..........................................................39
- 4. Determinants of NFCs, All countries ...................................................................................40
- 5. Determinants of NFC deposits, Middle Income countries ...................................................41
- 6. Determinants of NFC deposits, High Income Countries .....................................................42
- 7. Determinants of NFC Deposits, Low Income Countries .....................................................43
- 8. Summary Statistics...............................................................................................................44
- 9a. Determinants of M0 Growth: Capital flows given by BIS 7B ...........................................45
- 9b. Determinants of M0 Growth: Capital flows given by BIS 7B+12D .................................46
- 10a. Determinants of M2 Growth: Capital flows given by BIS 7B .........................................47
- 10b. Determinants of M2 Growth: Capital flows given by BIS 7B+12D ...............................48

### Figures
- 1. Non-bank firm as surrogate intermediary ..............................................................................5
- 2. International debt securities outstanding (all borrowers) of developing countries by nationalilty and residence of the borrower ..................................................................................7
- 3. NFC international debt securities outstanding of developing economies by nationality of issuer ..........................................................................................................................................8
- 4. Global Broad Money (M2) and U.S. Dollar Global Liquidity GL of global NFC deposits (in Trillion of US$) 2002Q1–2013Q2 .....................................................................................11
- 5. Global Broad Money (M2) and U.S. Dollar Global Liquidity GL of global NFC deposits: Annual Growth Rates, 2002Q1–2013Q2 .................................................................................11
- 6. Global liquidity variable in U.S. dollar, Euro, and Yen ......................................................12
- 7. Levels of U.S. Dollar Global Liquidity of NFC deposits for all countries (left axis) and U.S. Dollar Global Liquidity of NFC deposits for middle income countries (right axis), 2003Q1–2013Q2. ...........................................................................................................14
- 8. Annual growth rates of U.S. Dollar Global Liquidity of NFC deposits for all countries (left axis) and for middle income countries (right axis) 2003Q1–2013Q2. ....................................15
- 9. Average annual growth rates of NFC deposits in domestic currency for all countries and for middle income countries. ...................................................................................................16
- 10. Money stock and capital inflows due to financial activities of NFCs ...............................23

*Source: _wp1409 - References*

### 11. Evolution of Global Aggregates and Capital Flows, 1996–2011 ......................................26

### 11. Evolution of Global Aggregates and Capital Flows, 1996–2011

### Introduction
- Purpose: examine relationship between bank liability aggregates and global financial conditions, focusing on bank liabilities as counterparts to banking sector assets and as indicators of credit availability and financial vulnerability.
- Key hypotheses:
  - Rapid credit growth is an indicator of potential financial vulnerability; normalized measures such as credit to GDP ratio matter.
  - The money aggregate associated with claims of non‑financial corporations (NFCs) on the banking sector (denoted L) is closely correlated with financial conditions facing firms operating across borders and with global economic activity.
- Rationale for liability‑side measures:
  - Bank liabilities are more transparent and homogenous than assets; largely short term and close to market values.
  - Liabilities can be organized into core and non‑core with contrasting cyclical properties; non‑core liabilities are more procyclical.
  - NFC activity that straddles borders can create currency mismatches and carry‑trade like positions not captured well by residence‑based external debt statistics.
- Data source: standardized reporting forms (SRFs) submitted to the IMF and published in International Financial Statistics (IFS).

### First look at the data (definition and construction of aggregates)
- Definition:
  - L: sum of deposits of NFCs in the banking system (“other depository corporations (ODCs)”), including transferable and other deposits included in or excluded from broad money definitions.
  - GL (Global Liquidity): sum across countries of each country’s L converted into a common numeraire currency (e.g., U.S. dollar GL denoted ܮܩ
௎ௌ஽
).
- Data limitations and caveats:
  - SRFs do not provide consistent MMF or shadow‑banking claims; L includes only deposits in the regulated banking sector.
  - Quality of monetary statistics in SRFs is uneven; some series show discontinuities and unexplained jumps.
  - Some key countries do not submit full SRF monetary data.
- Currency (numeraire) matters: the U.S. dollar GL measure incorporates exchange rate movements of the U.S. dollar vis‑à‑vis other currencies; numeraire choice affects information content of GL.

### Descriptive patterns and key statistics (selected figures and observations)
- Time series patterns:
  - Until around 2004 GL did not exceed 15 percent of global broad money; since then NFC deposits became a larger proportion of global broad money, with a sharp fall during the 2008 crisis.
  - Figure references (series behavior described):
    - Global Broad Money (M2) and Global NFC Deposits plotted in Trillion U.S. dollars (2003Q1–2012Q4 axis scaling described).
    - Annual growth rates of GL vs global broad money show GL is highly procyclical (2002Q1–2013Q2).
    - Global liquidity measured in U.S. dollar, Euro, and Yen differ substantially; USD and JPY series are more similar.
- Middle income countries:
  - U.S. dollar GL for middle income countries is more procyclical than full sample; axes scaled so right axis equals 20 percent of left axis.
  - U.S. dollar GL for middle income countries declines close to 30 percent in 2009Q1 year‑on‑year.
- Domestic currency measurement:
  - Average annual growth rates of NFC deposits measured in domestic currency show less dramatic differences between full sample and middle income countries; NFC deposit growth remains positive at crisis trough when measured in domestic currency.
- Historical parallel: Japan in the 1980s saw NFCs acting as surrogate intermediaries—raising funds in capital markets and recycling into bank deposits.

### Global liquidity as an activity indicator (empirical panel results summary)
- Sample: quarterly panel regressions for 88 countries; dependent variables: country GDP growth (Table 1), imports (Table 2), exports (Table 3).
- Main empirical findings:
  - Lagged growth of U.S. dollar global NFC deposits (U.S. dollar GL) is strongly positively associated with real GDP growth across the panel (Table 1, columns 1 and 2).
  - U.S. dollar GL and Japanese yen GL are positively and strongly associated with country imports and exports; Euro GL is weakly positive or can have the “wrong” sign depending on specification (Tables 2 and 3).
  - Individual country NFC deposit growth is generally not significant for GDP growth and often enters with a negative sign for imports and exports.
  - Growth of U.S. broker dealer sector (proxy for global bank credit supply) enters with a positive and highly significant coefficient across regressions.
  - QE dummy: QE is a post‑crisis dummy; coefficient on QE is negative and significant for GDP growth, while interaction QE*VIX has a positive coefficient (mitigating growth when VIX is high). For imports/exports, QE*VIX often has a negative coefficient.
- Interpretation:
  - U.S. dollar (and JPY) GL capture funding currency dynamics that amplify quantity contractions through exchange rate movements; choice of numeraire matters for identifying funding currency status.
  - Global GL acts as a common global factor explaining co‑movement of activity across countries; country‑level NFC deposits reflect ability to take advantage of global ease of borrowing and may interact with pull/push factors.
  - Alternative interpretations remain possible (e.g., firms hoarding cash in expectation of investment opportunities).

### Determinants of NFC deposits (country‑level regressions summary)
- Objective: explain quarterly growth of individual country NFC deposits (in domestic currency) across income group subsamples (full sample, middle income, high income, low income; Tables 4–7).
- Key explanatory variable: “Capital Inflows” = growth in financial liabilities of “other sectors” in balance of payments (growth in debt security liabilities and loan liabilities of non‑financial corporations and other financial corporations). Note: includes NFCs and “other financial corporations” (OFCs).
- Main empirical findings:
  - For full sample (Table 4), capital flows variable is marginally significant both contemporaneously and lagged by four quarters.
  - Exchange rate impact: change in nominal exchange rate enters with a negative sign in some specifications, implying an appreciating domestic currency is associated with higher NFC deposit growth.
  - VIX index enters with a negative sign: higher financial market volatility associated with slower NFC deposit growth.
  - In some key specifications, capital inflows through non‑banks are positively and significantly associated with increases in NFC claims on the banking sector, consistent with NFCs acting as surrogate financial intermediaries.
- Limitations:
  - Capital Inflows measure is broader than NFC liabilities and includes non‑bank financial institutions; data availability constrains isolation of NFC activities.
  - Low R^2 values in several specifications and weak significance for some right‑hand variables in full sample regressions.

### Key policy‑relevant observations and implications (from text)
- U.S. dollar global liquidity performs well as an indicator of global financial conditions and is strongly associated with country‑level growth, exports, and imports; choice of funding currency for numeraire is critical.
- Liability‑side monetary aggregates (NFC deposits) can reveal global dimensions of financial conditions that are not apparent from residence‑based external debt measures.
- Data gaps and uneven SRF reporting limit measurement; reinstituting and improving collection of monetary aggregates and shadow‑banking related cash‑like holdings of firms would improve monitoring of global liquidity and vulnerabilities.
- Firm‑level consolidated balance sheet data would complement aggregate SRF data and better capture cross‑border NFC behavior.

*Source: Standardized Reporting Forms (SRFs) and International Financial Statistics (IFS) data and analysis as presented in the chapter "11. Evolution of Global Aggregates and Capital Flows, 1996–2011" (text, figures, and tables summarized).*

### Appendix 2). This subsample of countries consists of 48 of the 72 countries that constitute

### _wp1409 - Appendix 2). This subsample of countries consists of 48 of the 72 countries that constitute

### Middle-income subsample and NFC surrogate intermediation
- Sample composition:
  - Middle income subsample consists of 48 of the 72 countries that constitute the full sample.
  - The middle income group is the largest of the three income categories in the sample.
- Main interpretation:
  - Middle income countries are at an early stage of financial system development and typically do not have fully open capital markets or banking systems directly exposed to external financial conditions.
  - Non-financial corporations (NFCs) are conjectured to act as surrogate intermediaries channeling global liquidity into domestic financial markets; this role is expected to be more pronounced in middle income countries.
- Empirical evidence (Table 5 summary):
  - The four-quarter lagged capital inflow variable is positive and significant at the 1 percent level (columns 1, 2, and 3).
  - Exchange rate appreciation effect is negative and significant, consistent with the main hypothesis.
  - VIX in levels enters with a negative sign and is highly significant (columns 4, 5, and 6).
  - Column 3 includes additional regional dummies that classify countries according to region.
  - R-squared values (ܴଶ numbers) are higher for middle income countries relative to full sample panel regressions, though still quite low.
- Contrasts with other income groups:
  - Tables 6 and 7 (high income and low income subsamples) show capital inflows are not associated with increases in NFC deposits; capital inflow variables are insignificant in both subsamples.
  - For high income countries, the BIS bank lending variable (cross-border claims of BIS-reporting banks to non-banks from BIS locational banking statistics, Table 7B) enters with a positive sign and is significant at the 5 percent level in columns 4 and 5.

### NFC deposits, monetary aggregates, and global financial conditions
- Role of NFC deposit aggregates:
  - NFC deposit monetary aggregates convey information on the extent of capital inflows and surrogate intermediation by the NFC sector.
  - NFC deposits constitute only a small proportion of total broad money (M2) but are more procyclical compared to M2.
- Linkages between M0, M2, NFCs, and central bank operations (Figure 10 conceptual flow):
  - Exporting NFCs with U.S. dollar receivables hedge by incurring U.S. dollar liabilities (e.g., borrowing from international banks, issuing U.S. dollar-denominated securities such as kimchi bonds).
  - Exporting firm: sells USD and buys KRW in FX market → proceeds produce KRW financial assets → deposits in commercial banks (classified as corporate deposits) → captured in M2.
  - Central bank interventions to slow currency appreciation increase FX reserves and draw reserves from commercial banks, thereby creating narrow money, M0 (central bank reserves).
  - Sterilization possibilities are acknowledged but not depicted.
- Empirical counterparts and data sources:
  - BIS Locational Statistics, Table 7B is an empirical counterpart to NFC borrowing from global capital markets; coverage is geographically broad.
  - BIS securities database Table 12D provides outstanding debt securities issued by non-corporate borrowers.
  - Data sources include BIS, Fed, IFS, and authors’ estimates.

### Panel regression specifications for M0 and M2 determinants
- General setup:
  - Dependent variables: quarterly log difference (growth) of money stock—either M2 (broad money) or M0 (narrow money).
  - Panel regressions use quarterly data with country fixed effects and clustered standard errors at the country level.
  - A dynamic panel model using system GMM is also estimated to address endogeneity.
- Two specification forms presented (equations (1) and (2) in source):
  - ∆M_{c,t} = β0 + β1 Capital_Inflow_{c,t-1} + β2 ∆RER_{c,t-1} + β3 VIX_{t-1} + β4 ∆VIX_{t} + β5 ∆Interoffice_{t-1} + control_{c,t} + e_{c,t}
  - ∆M_{c,t} = β0 + β1 Capital_Inflow_{c,t-1} + β2 ∆RER_{c,t-1} + β3 VIX_{t-1} + β4 ∆VIX_{t} + β5 ∆Interoffice_{t-1} + β6 QE_dummy + control_{c,t} + e_{c,t}
- Variable definitions (preserved exactly as in source):
  - ∆M_{c,t} is the quarterly growth rate in money supply;
  - Capital_Inflow_{c,t-1} is the NFC sector capital inflow into country c in period t-1, as given by the quarterly log difference in the external claims of BIS reporting country banks on country c, with one quarter lag;
  - ∆RER_{c,t-1} is the quarterly log difference of the real exchange rate lagged by one quarter;
  - VIX_{t-1} is the average of the quarter log of the VIX index; ∆VIX_{t} is the contemporaneous log difference in the VIX from the previous quarter, not lagged;
  - ∆Interoffice_{t-1} is the growth in net Interoffice assets of foreign banks in the US from the quarter before, with one quarter lag;
  - QE_dummy is the dummy variable capturing Quantitative Easing period (QE 1: 2009 Q1-2010 Q1, QE 2: 2010 Q4-2011 Q2);
  - control_{c,t} includes GDP Growth, Inflation, and Debt-to-GDP ratio.

### Empirical findings for M0 and M2 (Tables 9a/9b and 10a/10b)
- M0 results (Tables 9a and 9b):
  - Table 9a (capital inflow defined by BIS7B only): capital inflow variable is positive and significant with a lagged dependent variable included.
  - Table 9b (capital inflow extended to include debt securities issuance): capital inflow is not significant.
  - Log VIX enters with a positive sign for M0, reflecting increased commercial bank reserves during the crisis period.
  - Exchange rate variable does not feature as an important determinant of M0.
  - Overall: results for M0 are somewhat mixed.
- M2 results (Tables 10a and 10b):
  - Table 10a (capital inflow = BIS7B): capital inflow variable is significant and positive; VIX enters with a negative sign, as predicted.
  - Table 10b (capital inflow extended to include debt securities growth): not all specifications show a significant effect of capital inflows.
  - Broad consistency with prior regressions focusing directly on NFC deposits.

### Time series patterns and graphical evidence (Figure 11 summary)
- Observed dynamics 1996–2011 (Figure 11):
  - M2 is procyclical; M0 displays some countercyclical features, especially around the time of the financial crisis.
  - U.S. dollar real effective exchange rate (REER) weakening in the middle 2000s coincided with an uptick in capital inflows as measured by the BIS 7B series.
  - Global aggregates plotted include Net Interoffice Assets (right axis), Global Capital Inflows (left axis), REER in the United States, Global M0 and M2 growth rates, Global trade and global GDP growth.
  - Note in figure legend: "All growth rates are annualized."

### Conclusions and policy implications
- Key conclusions:
  - Specialized bank liability aggregates (e.g., NFC deposits) can have information value as indicators of economic activity because of their sensitivity to global credit conditions.
  - Monetary aggregates, when adapted, can play an indicative role in:
    - Providing insights into credit condition patterns through liability-side aggregates of bank balance sheets.
    - Offering a systemic perspective on financial stability tied to banking sector procyclicality and vulnerability to reversals of capital flows.
- Core versus non-core liabilities distinction:
  - Core liabilities: funding drawn on during normal times and sourced mainly domestically (retail deposits of the household sector are a useful first conjecture).
  - Non-core liabilities: volatile funding sources tapped during lending booms (includes short-term foreign currency-denominated bank liabilities and wholesale corporate deposits).
  - The ratio of non-core to core liabilities is a reliable indicator of vulnerability to crises (reference to Hahm, Shin, and Shin (2013) in source).
- Adaptations recommended for monitoring and macroprudential policy:
  - For countries with open capital markets:
    - International capital flows into the banking sector are key indicators of financial vulnerability.
    - Short-term foreign currency-denominated bank liabilities can be treated as volatile non-core liabilities of the banking sector.
  - For countries with relatively closed financial systems:
    - Adapt conventional monetary aggregates by distinguishing who holds the claims: household retail deposits versus corporate deposits.
    - The distinction between retail deposits and NFC wholesale deposits is particularly important.
  - More generally:
    - Use the accounting principle that defines core versus non-core liabilities to guide classification of financial systems.
    - The relative size of non-core liabilities can serve as a monitoring tool reflecting the stage of the financial cycle, vulnerability to setbacks, and as an early warning indicator.
- Practical caveats:
  - Classification into core and non-core is not clear-cut (e.g., bank deposits of small owner-managed enterprises may resemble household deposits; large firms’ deposits may be wholesale).
  - Traditional monetary aggregates defined by legal form and transactional liquidity may be less effective as macroprudential tools without adaptation to claim-holder composition.
- Research implications:
  - Dynamic panel specifications could be appropriate to study causality effects given quarterly data and persistence of shocks; this is suggested as a follow-up research topic.

*Source: _wp1409 - Appendix 2). This subsample of countries consists of 48 of the 72 countries that constitute (IMF working paper content provided in the source PDF).*

### REFERENCES

### REFERENCES

### Cited Works
- Bank for International Settlements, 2010, “Funding patterns and liquidity management of internationally active banks”, CGFS paper 39, May 2010, http://www.bis.org/publ/cgfs39.htm
- Bank for International Settlements (2013) “Emerging market debt securities issuance in offshore centres” BIS Quarterly Review, September 2013, pp. 22–23 https://www.bis.org/publ/qtrpdf/r_qt1309w.htm
- Basel Committee on Banking Supervision, 2009, “Strengthening the Resilience of the Banking Sector,” December 2009, http://www.bis.org/publ/bcbs164.pdf
- Basel Committee on Banking Supervision, 2010, “International regulatory framework for banks (Basel III),” Bank for International Settlements, http://www.bis.org/bcbs/basel3.htm
- Borio, Claudio and Philip Lowe, 2002, “Asset Prices, Financial and Monetary Stability: Exploring the Nexus,” BIS Working Paper, No. 114, Basel: Bank for International Settlements, July.
- Borio, Claudio and Philip Lowe, 2004, “Securing sustainable price stability: should credit come back from the wilderness?” BIS Working Paper, No.157.
- Bruno, Valentina and Hyun Song Shin, 2012, “Capital Flows and the Risk-Taking Channel of Monetary Policy,” NBER Working Paper 18942, http://www.princeton.edu/~hsshin/www/capital_flows_risk-taking_channel.pdf
- Bruno, Valentina and Hyun Song Shin, 2013, “Cross-Border Banking and Global Liquidity,” http://www.princeton.edu/~hsshin/www/capital_flows_global_liquidity.pdf
- Chung, Kyuil, Hail Park, and Hyun Song Shin, 2012, “Mitigating Spillover Effects from Currency Hedging,” National Institute Economic Review, NIESR, London.
- Edge, Rochelle M., and Ralf R. Meisenzahl, 2011, “The Unreliability of Credit-to-GDP Ratio Gaps in Real Time: Implications for Countercyclical Capital Buffers,” International Journal of Central Banking, vol. 7, pp. 261–298.
- Filardo, Andrew and James Yetman, 2012, “The expansion of central bank balance sheets in emerging Asia: what are the risks?” BIS Quarterly Review, June 2012, pp. 47–63, http://www.bis.org/publ/qtrpdf/r_qt1206g.pdf
- Friedman, Benjamin, M., 1988, “Lessons on Monetary Policy from the 1980s,” Journal of Economic Perspectives, 2, pp. 51–72
- Friedman, Milton, 1956, “The Quantity Theory of Money - A Restatement,” in Studies in the Quantity Theory of Money (ed) Milton Friedman, Chicago University Press, Chicago.
- Hahm, Joon-Ho, Hyun Song Shin and Kwanho Shin, 2013, “Non-Core Bank Liabilities and Financial Vulnerability,” Journal of Money, Credit and Banking, 45(S1), pp. 3–36.
- Hattori, Masazumi, Hyun Song Shin, and Wataru Takahashi, 2009, “A Financial System Perspective on Japan’s Experience in the Late 1980s,” paper presented at the 16th Bank of Japan International Conference, May 2009, Bank of Japan IMES discussion paper, http://www.imes.boj.or.jp/english/publication/edps/2009/09-E-19.pdf
- Ivashina, Victoria and David Scharfstein, 2010, “Bank lending during the financial crisis of 2008,” Journal of Financial Economics, vol. 97, pp. 319–38.
- Kim, Hyun Jeong, Hyun Song Shin, and Jaeho Yun, 2013, “Monetary Aggregates and the Central Bank's Financial Stability Mandate,” International Journal of Central Banking, January 2013, pp. 69–107.
- McCauley, Robert, Christian Upper and Agustín Villar (2013) “Emerging market debt securities issuance in offshore centres” BIS Quarterly Review, September, pp. 22–23.
- McKinnon, Ronald I., 1982, “Currency Substitution and Instability in the World Dollar Standard,” American Economic Review, Vol. 72, No. 3, pp. 320–33.
- Shin, Hyun Song and Laura Yi Zhao, 2013, “Non-financial Firms as Surrogate Intermediaries” working paper, Princeton University.
- Turner, Philip, 2013, “The global long-term interest rate, financial risks and policy choices in EMEs”, paper for the Inter-American Development Bank Meeting of Chief Economists of Central Banks and Finance Ministries, Washington DC, October 2013.

### Appendix I. Data Description and Regressions Results — Variable Definitions and Sources
- NFC deposits: The sum of transferable and other deposits of public and other (private) non-financial corporations to other depository corporations (ODCs), included in and excluded from broad money, national and foreign currency — Other Depository Corporations Survey 2SG for the International Financial statistics (IFS), IMF, as reported by the country authorities
- Capital inflows: Debt securities and loans of other sectors (non-financial corporations and other financial corporations), gross liabilities — Balance of Payments and international Investment Position (compiled by the sixth edition methodology, BPM6), Statistics Department (STA), IMF, as reported by country authorities
- Exchange rate: Nominal exchange rate, end-of-period — Central bank survey 1SG for the IFS, IMF, as reported by country authorities
- VIX: Chicago Board Options exchange Market Volatility Index, the implied volatility of S&P 500 index options; average — Bloomberg
- Interoffice: Growth in net Interoffice assets of foreign banks in the United States — Federal Reserve Board (Fed) website
- QE: Dummy variable capturing Quantitative Easing periods (QE1: 2009Q1 – 2010Q1; QE2: 2010Q4 – 2011Q2) — Constructed by authors
- GDP growth: Real GDP growth, annual — National Accounts Database, STA/IMF, as reported by country authorities
- Inflation: Annual percentage change of the CPI, end of period — National accounts Database, STA/IMF, as reported by country authorities
- Debt/GDP ratio: Total external debt to GDP ratio — World Economic Outlook (WEO), IMF
- External loans to non-banks (BIS): External loans and deposits of reporting banks vis-à-vis the non-bank sector — Bank for International Settlements (BIS)
- Spread: Spread between average deposit rate and BofA Merrill Lynch US High Yield BB US Effective Yield — FRED database http://research.stlouisfed.org/fred2/
- Broad money: Broad money liabilities — Depository corporations survey 3SG for IFS
- Global NFC deposits: Sum of the non-financial corporate deposits in USD (EUR, JPY) — Other Depository Corporations Survey 2SG for the International Financial statistics (IFS), IMF, as reported by the country authorities
- Export: Export of goods and non-factor services (in U.S. dollars) — Balance of Payments and international Investment Position (compiled by the sixth edition methodology, BPM6), Statistics Department (STA), IMF, as reported by country authorities
- Import: Import of goods and non-factor services (in U.S. dollars) — Balance of Payments and international Investment Position (compiled by the sixth edition methodology, BPM6), Statistics Department (STA), IMF, as reported by country authorities
- Sovereign debt issuance: Sovereign and central bank debt securities — Balance of Payments and international Investment Position (compiled by the sixth edition methodology, BPM6), Statistics Department (STA), IMF, as reported by country authorities

### Appendix II. Composition of SRF Submitting Countries by Income Group (used in empirical analysis)
- The construction of the global NFC aggregate includes data from the following countries that submit the Standardized Reporting Form (SRF) to the IMF. The classification of the income group follows the World Bank classification.

- LOW INCOME COUNTRIES (US$1,035 or less):
  - Afghanistan, Bangladesh, Burundi, Cambodia, Central African Republic, Chad, Comoros, Congo, DR, Eritrea, Ethiopia, The Gambia, Haiti, Kenya, Mali, Mozambique, Myanmar, Namibia, Nepal, Sierra Leone, Tajikistan, Tanzania, Uganda

- MIDDLE INCOME COUNTRIES (US$1,036 to US$12,615):
  - Albania, Algeria, Anguilla, Armenia, Azerbaijan, Belarus, Belize, Bhutan, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Cameroon, Cape Verde, Colombia, Congo Rep., Costa Rica, Dominica, Dominican Republic, Ecuador, Egypt, El Salvador, Fiji, Gabon, Georgia, Ghana, Grenada, Guatemala, Guyana, Honduras, Hungary, Indonesia, Iraq, Jamaica, Kazakhstan, Kosovo, Lesotho, Macedonia FYR, Malaysia, Maldives, Mauritius, Mexico, Moldova, Mongolia, Morocco, Nicaragua, Nigeria, Pakistan, Panama, Papua New Guinea, Paraguay, Philippines, Romania, Samoa, Sao Tome and Principe, Serbia, Seychelles, Solomon Islands, South Africa, St. Lucia, St. Vincent and the Grenadines, Sudan, Suriname, Swaziland, Syrian Arab Republic, Thailand, Timor-Leste, Tonga, Turkey, Ukraine, Vanuatu, Venezuela, West Bank and Gaza, Zambia

- HIGH INCOME COUNTRIES (US$12,616 or more):
  - Antigua and Barbuda, Australia, Barbados, Bermuda, Brunei Darussalam, Canada, Chile, Croatia, Czech Republic, Denmark, Equatorial Guinea, Euro Area, Iceland, Israel, Japan, Korea, Kuwait, Latvia, Lithuania, Macao SAR, Malta, Oman, Poland, Qatar, St. Kitts and Nevis, Sweden, Trinidad and Tobago, United States, Uruguay

### Regression Framework and Sample
- Sample period for panel regressions: 2001Q4–2013Q1
- Frequency: quarterly
- Regression types: unbalanced panel regressions
- Dependent variables reported in tables:
  - Table 1: real GDP growth
  - Table 2: import of goods and services
  - Table 3: export of goods and services
  - Table 4: log-difference of total NFC deposits (all countries)
- Tables report observations, R-squared, number of country_code, and robust p-values in brackets. Significance notation used: *** p<0.01, ** p<0.05, * p<0.1

### Selected Table Metadata and Key Values (preserved exactly as in source)
- Table 1 (Global Money Stock as Activity Indicator for real GDP growth):
  - Sample period: 2001Q4–2013Q1
  - Observations: 3,505; 3,130; 3,505; 3,130; 3,505; 3,130 (as reported across regressions)
  - R-squared: 0.1030; 0.1500; 0.0960; 0.1400; 0.1040; 0.150
  - Number of country_code: 88; 83; 88; 83; 88; 83
  - Examples of reported coefficients (with robust pval in brackets):
    - Dlog(Global NFC in USD) (t-1): 1,093*** [1.39e-07]; 1,138*** [4.89e-09]
    - Dlog(Global NFC in JPY) (t-1): 1,021*** [7.84e-09]; 1,011*** [1.95e-08]
    - Dlog(Broad Money): 0.0521** [0.0144]; 0.0562** [0.0171]; 0.0599*** [0.00705]; 0.0661*** [0.00677]; 0.0563*** [0.00876]; 0.0612*** [0.00957]
    - QE: -0.0277*** [2.58e-09]; -0.0282*** [2.14e-10]; -0.0249*** [7.83e-09]; -0.0253*** [8.26e-10]; -0.0216*** [1.44e-07]; -0.0220*** [1.17e-08]
    - Constant: 0.0479*** [0]; 0.0467*** [0]; 0.0496*** [0]; 0.0483*** [0]; 0.0486*** [0]; 0.0474*** [0]

- Table 2 (Global Money Stock as Activity Indicator of Imports):
  - Observations: 3,239; 3,213; 3,239; 3,213; 3,239; 3,213
  - R-squared: 0.0680; 0.0770; 0.0280; 0.0520; 0.0520; 0.066
  - Number of country_code: 85 (across regressions)
  - Examples of reported coefficients:
    - Dlog(Global NFC in USD) (t-1): 8,230*** [0]; 6,757*** [0]
    - Dlog(Global NFC in JPY) (t-1): 5,266*** [0]; 4,482*** [0]
    - Dlog(NFC) (t-1): -0.0461** [0.0122]; -0.0478*** [0.00994]; -0.0431** [0.0206]; -0.0458** [0.0139]; -0.0442** [0.0157]; -0.0464** [0.0116]
    - QE: -0.0768*** [6.57e-08]; -0.0617*** [9.43e-06]; -0.0644*** [1.01e-06]; -0.0459*** [0.000284]; -0.0376*** [0.00189]; -0.0301** [0.0114]

- Table 3 (Global Money Stock as Activity Indicator of Export):
  - Observations: 3,239; 3,213; 3,239; 3,213; 3,239; 3,213
  - R-squared: 0.0520; 0.0680; 0.0340; 0.0580; 0.0470; 0.064
  - Number of country_code: 85 (across regressions)
  - Examples of reported coefficients:
    - Dlog(Global NFC in USD) (t-1): 6,296*** [0]; 4,887*** [5.51e-08]
    - Dlog(Global NFC in JPY) (t-1): 4,406*** [5.63e-10]; 3,279*** [5.47e-06]
    - Dlog(NFC) (t-1): -0.0542** [0.0165]; -0.0513** [0.0241]; -0.0522** [0.0192]; -0.0493** [0.0288]; -0.0528** [0.0168]; -0.0504** [0.0239]
    - QE: -0.0714*** [8.12e-10]; -0.0588*** [2.49e-07]; -0.0603*** [3.27e-08]; -0.0481*** [6.35e-06]; -0.0395*** [0.000130]; -0.0355*** [0.000465]

- Notes on statistical reporting:
  - Robust pval in brackets
  - Significance markers: *** p<0.01, ** p<0.05, * p<0.1

*IMF Working Paper — REFERENCES and Appendices (extracted from _wp1409 - REFERENCES).*

### Appendix 1.

### Appendix 1.

### Determinants of NFC deposits — Middle Income countries (Table 5)
- Sample period and frequency: 2001Q4–2013Q1, quarterly.
- Dependent variable: log-difference of total NFC deposits in banks for middle-income countries.
- Notable coefficients (robust p-value in brackets; significance stars as reported):
  - Capital inflows (t-4): 0.0108*** [0.00328]; 0.0108*** [0.00311]; 0.0102*** [0.00621].
  - ∆ Nominal Exchange rate (t-1): -2.03e-05* [0.0599]; -2.08e-05** [0.0298]; -2.71e-05* [0.0762].
  - Log(VIX) (t-1): -0.0214*** [0.00604]; -0.0200** [0.0126]; -0.0201** [0.0190].
  - Dlog(VIX) (t-1): 0.0233* [0.0949]; 0.0291* [0.0554]; 0.0227 [0.101].
  - Growth of interoffice (t-1): 0.0120* [0.0509] (one specification).
  - QE * VIX (t-1): -20.84* [0.0932] (one specification).
  - QE * VIX (CIS): -18.19*** [0.00581].
  - QE * VIX (SSA): 9.139*** [0.00687].
- Observations: 535 (panel regressions reported variously with Observations 535 and 1,704 in alternative columns).
- R-squared: 0.029; 0.034; 0.044 (reported across columns).
- Number of country_code: 38; 48 (across specifications).

### Determinants of NFC deposits — High Income countries (Table 6)
- Sample period and frequency: 2001Q4–2013Q1, quarterly.
- Dependent variable: log-difference of total NFC deposits in banks for high-income countries.
- Notable coefficients:
  - Dlog(BIS external loans nonbanks) (t-1): 0.0659* [0.0694] (one specification).
  - Dlog(BIS external loans nonbanks) (t-4): 0.125** [0.0200]; 0.108** [0.0370]; 0.104* [0.0520].
  - QE(t-1): -0.0746* [0.0643] (one specification, negative).
  - QE * VIX (t-1): 341.22 [0.287] (one specification with large coefficient, not significant).
  - Inflation (t-1): 0.0116* [0.0898] in one specification.
- Observations: 124 (columns with panel observations reported as 124 and 309).
- R-squared: 0.042; 0.067; 0.056; 0.035; 0.042; 0.043 (across specifications).
- Number of country_code: 9 (repeated in table).

### Determinants of NFC deposits — Low Income countries (Table 7)
- Sample period and frequency: 2001Q4–2013Q1, quarterly.
- Dependent variable: log-difference of total NFC deposits in banks for low-income countries.
- Notable coefficients:
  - Dlog(VIX) (t-1): 0.0987*** [0.00593] (one specification); other columns show smaller, insignificant values.
  - QE(t-1): 0.0554* [0.0688] in one specification (positive).
  - Δ Debt/GDP (t-1): -9.961* [0.0659]; -13.32*** [0.00861]; -13.79** [0.0147] (negative and significant in fixed-effect specifications).
- Observations: 353; 472 (across columns).
- R-squared: 0.172; 0.199; 0.196 (for fixed-effect columns); 0.004; 0.004; 0.005 (other columns).
- Number of country_code: 7; 14 (across specifications).

### Summary statistics (Table 8)
- Global variables:
  - VIX (Quarter): Obs = 70; Mean = 3.0161; Std. Dev. = 0.3371; Min = 2.4006; Max = 4.0707.
  - ΔVIX (Quarter): Obs = 69; Mean = 0.0074; Std. Dev. = 0.2134; Min = -0.4978; Max = 0.8491.
  - Δ Interoffice (Quarter): Obs = 65; Mean = -0.0001; Std. Dev. = 1.1145; Min = -4.1770; Max = 5.2829.
- Local variables (selected):
  - ΔM0 (Quarter): Obs = 3,011; Mean = 0.0269; Std. Dev. = 0.1602; Min = -3.1091; Max = 2.9787.
  - ΔM2 (Quarter): Obs = 2,977; Mean = 0.0308; Std. Dev. = 0.1320; Min = -2.3519; Max = 2.6670.
  - Capital Inflow (Quarter): Obs = 2,970; Mean = 0.0235; Std. Dev. = 0.1100; Min = -0.7631; Max = 0.9285.
  - Δ RER (Quarter): Obs = 2,917; Mean = -0.0006; Std. Dev. = 0.0661; Min = -0.5101; Max = 1.0309.
  - Real GDP Growth (Annual): Obs = 810; Mean = 0.0326; Std. Dev. = 0.0376; Min = -0.1480; Max = 0.1420.
  - Inflation (Annual): Obs = 802; Mean = 0.0679; Std. Dev. = 0.2507; Min = -0.1285; Max = 5.4921.
  - Δ Debt to GDP (Annual): Obs = 700; Mean = 0.0043; Std. Dev. = 0.0773; Min = -0.3991; Max = 1.1137.

### Determinants of M0 growth (Tables 9a and 9b)
- Panel regressions (quarterly) of M0 growth with capital flows measured by BIS 7B (Table 9a) and BIS 7B+12D (Table 9b).
- Sample periods reported:
  - Fixed effects: Sample period: 1995Q1 - 2012Q2.
  - Dynamic (System GMM): Sample period: 2000Q1 - 2012Q2.
- Notable coefficients (selected and preserved exactly):
  - M0 Growth(t-1): -0.2986*** [0.0650]; -0.2987*** [0.0649]; -0.2143*** [0.0808]; -0.2160*** [0.0807].
  - Capital Inflow(t-1): 0.0251** [0.0117]; 0.0252* [0.0117]; 0.0269** [0.0127]; 0.0266** [0.0127]; 0.0272* [0.0148]; 0.0273* [0.0149]; 0.0294* [0.0164]; 0.0288* [0.0164].
  - VIX(t-1): 0.0144*** [0.0035]; 0.0158*** [0.0036]; and in other columns 0.0011 [0.0046]; 0.0015 [0.0045].
  - △VIX: 0.0301** [0.0125]; 0.0293* [0.0130]; 0.0324** [0.0136]; 0.0267* [0.0148] (selected columns).
  - △Interoffice(t-1): 0.0075*** [0.0022]; 0.0075*** [0.0022]; 0.0054** [0.0022]; 0.0051** [0.0021]; 0.0082*** [0.0022]; 0.0082*** [0.0022]; 0.0078*** [0.0021]; 0.0072*** [0.0019].
  - QE: -0.0159* [0.0086] (one System GMM specification).
  - Real GDP Growth(t-1): 0.1186** [0.0510]; 0.1118* [0.0568]; 0.2696*** [0.0693]; 0.2281*** [0.0792].
  - Inflation(t-1): 0.1025*** [0.0359]; 0.1010*** [0.0357]; 0.1995*** [0.0438]; 0.1876*** [0.0403].
- Observations and country counts (selected):
  - Observations: 2,646; 2,646; 2,462; 2,640; 2,640; 2,134; 2,134; 2,134 (across columns).
  - # of Countries: 45 (repeated across columns).
- AR tests:
  - AR(1) p-value: 0.0001; 0.0001; 0.0006; 0.0006 (across dynamic specifications).
  - AR(2) p-value: 0.0603; 0.0670; 0.9818; 0.9673 (reported).

### Determinants of M2 growth (Tables 10a and 10b)
- Panel regressions (quarterly) of M2 growth with capital flows measured by BIS 7B (Table 10a) and BIS 7B+12D (Table 10b).
- Sample periods reported:
  - Fixed effects: Sample period: 1995Q1 - 2012Q2.
  - Dynamic (System GMM): Sample period: 2000Q1 - 2012Q2.
- Notable coefficients (selected and preserved exactly):
  - M2 Growth(t-1): -0.3838*** [0.0800]; -0.3831*** [0.0799]; -0.2997*** [0.1151]; -0.2995*** [0.1140] (and similar in Table 10b: -0.3844***; -0.3837***; -0.3003***; -0.3001***).
  - Capital Inflow(t-1): 0.0145* [0.0082]; 0.0146* [0.0082]; 0.0178* [0.0101]; 0.0177* [0.0100]; 0.0199* [0.0106]; 0.0200* [0.0106]; 0.0336** [0.0164]; 0.0338** [0.0164] (Table 10a); Table 10b shows smaller but similar positive coefficients, including 0.0349* [0.0199]; 0.0360* [0.0202] in some columns.
  - △RER(t-1): negative and sometimes significant: -0.0420* [0.0245]; -0.0431* [0.0250]; -0.0312* [0.0170]; -0.0321* [0.0172] (selected columns).
  - VIX(t-1): mixed signs and significance; e.g., -0.0072** [0.0030]; -0.0064* [0.0030]; 0.0151** [0.0029]; 0.0163*** [0.0031]; Table 10b similar pattern with -0.0075** [0.0030]; -0.0067** [0.0030]; 0.0151*** [0.0029]; 0.0162*** [0.0031].
  - QE: -0.0127*** [0.0045] (one specification negative and significant); other columns show QE coefficients like -0.0037 [0.0025]; -0.0095** [0.0037].
  - Real GDP Growth(t-1): strongly positive and significant in multiple specifications, e.g., 0.1642*** [0.0262]; 0.1518*** [0.0288]; 0.4003*** [0.0603]; 0.3648*** [0.0579].
  - Inflation(t-1): positive and often significant, e.g., 0.0805*** [0.0181]; 0.0778*** [0.0186]; 0.1397*** [0.0356]; 0.1296*** [0.0347].
- Observations and country counts (selected):
  - Observations: 2,623; 2,623; 2,326; 2,326; 2,614; 2,614; 2,134; 2,134 (across columns).
  - # of Countries: 45 (repeated across columns).
- AR tests:
  - AR(1) p-value: 0.0067; 0.0067; 0.0124; 0.0126 (across dynamic specifications).
  - AR(2) p-value: 0.2079; 0.2044; 0.6865; 0.6667 (reported).

*Italic: Source — Appendix 1, Tables and regression outputs as provided in the supplied content unit.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1409.pdf_