Skip to main content
Log in

A macroeconomic analysis of foreign assets and foreign liabilities

  • Published:
Journal of Economics and Finance Aims and scope Submit manuscript

Abstract

Although an increase in foreign assets and a decrease in foreign liabilities both increase a nation’s net foreign assets (NFA), they have alternative macroeconomic transmission mechanisms: while an increase in foreign assets is expansionary, the effect of a decrease in foreign liabilities is mixed due to the asymmetry between its income effect and wealth effect on aggregate demand. It is the relative strengths of the NFA’s wealth effect and income effect that determine the existence and natures of a saddle-point equilibrium in the NFA-real balance space as well as its comparative statics. The cointegration analysis suggests that in the 1990s, foreign liabilities bear more weight than foreign assets in the US NFA movement whereas the opposite holds for the case of Japan; therefore, correcting NFA imbalances calls for accelerated money growth and fiscal expenditure pruning in the U.S. but for the opposite policy responses in Japan.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. For the classical work on the wealth effect, see Tobin (1958, 1969).

  2. A recent study (see Albuquerque 2003) offers a new risk-sharing perspective to show that foreign direct investment, being less volatile than other financial flows, plays an important role in aggregate supply.

  3. It is assumed that national capital wealth, K d  + NFA, is nonnegative, that is, a negative NFA position, if exists, cannot make national capital wealth negative.

  4. For simplicity, government taxes are abstracted away in the model.

  5. Given the simplified assumption that inflation rate equals the rate of exchange rate depreciation in this paper, the outcome obtained in this case echoes Lane and Milesi-Ferretti (2004) in which the cross-country data on real exchange rates and net external positions suggest that on average deteriorating net external liabilities are associated with depreciated real exchange rates.

  6. Note that although the stable path is often considered as an economically sensible result if people behavior is governed by rational expectation or people can at least learn to form an optimal forecast in a plausible way (see Brock 1975; Evans 1985), Gray (1984) challenges the common practice of arbitrarily choosing the stable path as the unique equilibrium solution and shows that it is neither justifiable nor desirable to rule out a multiplicity of non-steady-state solutions.

  7. The presence of the income effect makes the \( {\text{N\ifmmode\expandafter\dot\else\expandafter\.\fi{F}A}} = 0 \) schedule to become flatter compared with the other case in which the income effect is either absent or weak.

  8. Although a classical work by Sidrauski (1967) favors the superneutrality, a number of studies appear not to be supportive (e.g., Romer 1986; Abel 1987; and Fisher and Seater 1993). Bullard and Keating (1994), however, presents mixed evidence from a sample of the postwar economies: some low-inflation countries exhibit nonsuperneutrality while most of the others show the opposite.

  9. In the June 2002 issue of International Financial Statistics, a nation’s net foreign assets equal to the sum of foreign assets held by both the monetary authority (series 11) and deposit money banks (series 21) less the sum of foreign liabilities in both the monetary authority (series 16c) and deposit money banks (series 26c).

  10. The finding on the policy implications to Japan contrasts with some mainstream policy perspectives of the Japanese economy, such as Krugman (2000).

Reference

  • Abel AB (1987) Optimal money growth. J Monetary Econ 19(3):437–450

    Article  Google Scholar 

  • Agenor P-R (1998) Capital inflows, external shocks, and the real exchange rate. J Int Money Financ 17(5):713–740

    Article  Google Scholar 

  • Albuquerque R (2003) The composition of international capital flows: risk sharing through foreign direct investment. J Int Econ 61(2):353–383

    Article  Google Scholar 

  • Borensztein E, de Gregorio J, Lee J-W (1994) How does foreign direct investment affect economic growth? International monetary fund working paper: WP/94/110

  • Brock W (1975) A simple perfect foresight monetary model. J Monetary Econ 1(2):133–150

    Article  Google Scholar 

  • Bullard J, Keating J (1994) Superneutrality in postwar economies. Federal reserve bank of St. Louis working paper, 94–011A

  • Evans G (1985) Expectational stability and the multiple equilibria problem in linear rational expectations models. Quart J Econ 100(4):1217–1233

    Article  Google Scholar 

  • Fisher ME, Seater JJ (1993) Long-run neutrality and superneutrality in an ARIMA framework. Am Econ Rev 83(3):402–415

    Google Scholar 

  • Friedman BM (1986) Implications of the U.S. net capital inflow. In: Hafer RW (ed) How open is the U.S. Economy? Heath, Lexington books, Lexington, MA, pp 137–161

    Google Scholar 

  • Gray JA (1984) Dynamic instability in rational expectations models: an attempt to clarify. Int Econ Rev 25(1):93–122

    Article  Google Scholar 

  • Howard D (1989) Implications of the U.S. current account deficit. J Econ Perspect 3(4):153–165

    Google Scholar 

  • Krugman P (1979) A model of balance-of-payments crises. J Money, Credit, Banking 11(3):311–325

    Article  Google Scholar 

  • Krugman P (2000) Thinking about the liquidity trap. J Japan Int Econ 14(4):221–237

    Article  Google Scholar 

  • Lane P, Milesi-Ferretti GM (1999) The external wealth of nations: measures of foreign assets and liabilities for industrial and developing countries. International monetary fund working paper, WP/99/115

  • Lane P, Milesi-Ferretti GM (2004) The transfer problem revisited: net foreign assets and real exchange rates. Rev Econ Stat 86(4):841–857

    Article  Google Scholar 

  • Lane P, Milesi-Ferretti GM (2006) The external wealth of nations Mark II: revised and extended estimates of foreign assets and liabilities. International monetary fund working paper WP/06/69

  • Romer D (1986) A simple general equilibrium version of the baumol-tobin model. Quart J Econ 101(4):663–686

    Article  Google Scholar 

  • Selaive J, Tuesta V (2003) Net foreign assets and imperfect pass-through: the consumption real exchange rate anomaly. International finance discussion papers 764. Board of governors of the federal reserve system, Washington, DC

    Google Scholar 

  • Sidrauski M (1967) Rational choice and patterns of growth in a monetary economy. Am Econ Rev 57(2):534–544

    Google Scholar 

  • Tobin J (1958) Liquidity preference as behavior toward risk. Rev Econ Stud February, 65–86

  • Tobin J (1969) A general equilibrium approach to monetary theory. J Money Credit Banking 1(1):15–29

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ying Wu.

Additional information

The author has benefited greatly from an anonymous JEF referee’ helpful comments on an earlier verasion of the paper. A previous summer research grant from Franklin P. Perdue School of Salisbury University is gratefully acknowledged.

Appendices

Appendix 1 Determination of the saddle-point equilibrium

First, consider the case in which dK* d  ≠ 0 and dK f    = 0. Let J 1 be the Jacobian matrix of Eq. 6 for the case in which foreign assets drive changes in NFA. According to Eq. 6, the determinant of J 1 can be written as

$$ {\left| {J_{1} } \right|} = m{C}\ifmmode{'}\else$'$\fi_{1} {\left( {{h}\ifmmode{'}\else$'$\fi_{1} - {h}\ifmmode{'}\else$'$\fi_{2} } \right)} < 0, $$
(A1)

where \({h}\ifmmode{'}\else$'$\fi_{1} = L^{{ - 1}}_{1} \prime \frac{{K_{d} + {\text{NFA}}}}{{{\left( {K_{d} + {\text{NFA}} + m} \right)}^{2} }} < 0\) and \({h}\ifmmode{'}\else$'$\fi_{2} = L^{{ - 1}}_{1} \prime {\left[ { - \frac{m}{{{\left( {K_{d} + {\text{NFA}} + m} \right)}^{2} }}} \right]} > 0\).

The sign for the trace of J 1 is undetermined since

$$ Tr{\left( {J_{1} } \right)} = - C'_{\kern2pt1} - mh'_{\kern2pt1} . $$
(A2)

It follows that the system has a saddle-point equilibrium when changes in NFA totally stem from foreign assets only.

Next, turning to the case in which dK f    ≠ 0 and dK* d  = 0, the determinant of the Jacobian matrix J 2 in Eq. 9 is

$$ {\left| {J_{2} } \right|} = m{h}\ifmmode{'}\else$'$\fi_{2} {\left[ {{F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{2} } \right)} + {C}\ifmmode{'}\else$'$\fi_{1} } \right]}{\left[ { - \frac{{{C}\ifmmode{'}\else$'$\fi_{1} }} {{{F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{2} } \right)} + {C}\ifmmode{'}\else$'$\fi_{1} }} - {\left( { - \frac{{{h}\ifmmode{'}\else$'$\fi_{1} }} {{{h}\ifmmode{'}\else$'$\fi_{2} }}} \right)}} \right]}, $$
(A3)

where h1 remains the same as above but \(h^{\prime } _{2} = - {\left[ {L^{{ - 1}}_{2} \prime {F}\ifmmode{'}\else$'$\fi + L^{{ - 1}}_{1} \prime \frac{m}{{{\left( {K_{d} + {\text{NFA}} + m} \right)}^{2} }}} \right]} < 0\) if and only if the first term in the bracket (the income effect) is greater than the second term (the wealth effect). ∣J 2∣is negative either when h′ 2 > 0, or when h2 < 0 and the term in the second bracket of Eq. A3 has a positive sign, which actually means that the \({\mathop m\limits^. } = 0\) schedule is steeper than the \({\text{N}}{\mathop {\text{F}}\limits^{\text{.}} }{\text{A}} = 0\) schedule. Additionally, the trace of J 2 remains unsigned:

$$ Tr{\left( {J_{2} } \right)} = {\left[ { - {F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{2} } \right)} - {C}\ifmmode{'}\else$'$\fi_{1} } \right]} - m{h}\ifmmode{'}\else$'$\fi_{1} . $$
(A4)

Therefore, under these conditions, the system continues to have a saddle-point equilibrium.

Appendix 2 Slope of the stable branch in the saddle-point equilibrium

Again, first consider the case in which dK* d  ≠ 0 and dK f    = 0. The characteristic equation associated with the dynamic system (6) has two characteristic roots:

$$ \lambda {\kern 1pt} {\kern 1pt} _{1} = - \frac{1} {2}{\left( {{C}\ifmmode{'}\else$'$\fi_{1} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)} - \frac{1} {2}{\sqrt {{\left( {{C}\ifmmode{'}\else$'$\fi_{1} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)}^{2} - 4{\left( {{C}\ifmmode{'}\else$'$\fi_{1} m{h}\ifmmode{'}\else$'$\fi_{1} - {C}\ifmmode{'}\else$'$\fi_{1} m{h}\ifmmode{'}\else$'$\fi_{2} } \right)}} } < 0, $$

and

$$ \lambda {\kern 1pt} {\kern 1pt} _{2} = - \frac{1} {2}{\left( {{C}\ifmmode{'}\else$'$\fi_{1} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)} + \frac{1} {2}{\sqrt {{\left( {{C}\ifmmode{'}\else$'$\fi_{1} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)}^{2} - 4{\left( {{C}\ifmmode{'}\else$'$\fi_{1} m{h}\ifmmode{'}\else$'$\fi_{1} - {C}\ifmmode{'}\else$'$\fi_{1} m{h}\ifmmode{'}\else$'$\fi_{2} } \right)}} } > 0. $$

A solution must assume the form

$$ {\left( {\begin{array}{*{20}c} {{{\text{NFA}} - {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} \\ {{m - \ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} \\ \end{array} } \right)} = {\left( {\begin{array}{*{20}c} {{b_{{11}} }} & {{b_{{12}} }} \\ {{b_{{21}} }} & {{b_{{22}} }} \\ \end{array} } \right)}{\left( {\begin{array}{*{20}c} {{e^{{\lambda _{1} t}} }} \\ {{e^{{\lambda _{2} t}} }} \\ \end{array} } \right)}, $$
(A5)

where b ij are determined by the characteristic vectors of the Jacobian matrix in Eq. 6 and the initial values of NFA and m. For the system to be convergent to its steady state, the initial conditions must be such that b 12 = b 22 = 0, so NFA and m converge exponentially to \(\overline{{{\text{NFA}}}} \) and \(\overline{m} \). Then, from Eq. A5, we have \({\mathop m\limits^. } = \lambda {\kern 1pt} _{1} {\left( {m - \ifmmode\expandafter\bar\else\expandafter\=\fi{m}} \right)}\), which can be combined with the equation for \({\mathop m\limits^. }\) in Eq. 6 to yield

$$ {\text{NFA}} - {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}} = - \frac{{\lambda {\kern 1pt} _{1} + m{h}\ifmmode{'}\else$'$\fi_{1} }} {{m{h}\ifmmode{'}\else$'$\fi_{2} }}{\left( {m - \ifmmode\expandafter\bar\else\expandafter\=\fi{m}} \right)}. $$
(A6)

Eq. A6 defines the positively sloped stable branch for the case in which changes in NFA are produced by changes in foreign assets only.

Likewise, for the other case in which dK f  ≠ 0 and dK* d  = 0, the characteristic equation associated with the dynamic system (9) has the following two characteristic roots:

$$\mu {\kern 1pt} _{1} = - \frac{1}{2}{\left( {\phi {\kern 1pt} {\kern 1pt} + mh^{\prime } _{1} } \right)} - \frac{1}{2}{\sqrt {{\left( {\phi {\kern 1pt} {\kern 1pt} {\kern 1pt} + mh^{\prime } _{1} } \right)}^{2} - 4{\left( {\phi {\kern 1pt} {\kern 1pt} {\kern 1pt} mh^{\prime } _{1} - {C}\ifmmode{'}\else$'$\fi_{1} mh^{\prime } _{2} } \right)}} } < 0,$$

where \( \phi \equiv {F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{1} } \right)} + {C}\ifmmode{'}\else$'$\fi_{1} \), and

$$ \mu {\kern 1pt} _{2} = - \frac{1} {2}{\left( {\phi {\kern 1pt} {\kern 1pt} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)} + \frac{1} {2}{\sqrt {{\left( {\phi {\kern 1pt} {\kern 1pt} {\kern 1pt} + m{h}\ifmmode{'}\else$'$\fi_{1} } \right)}^{2} - 4{\left( {\phi {\kern 1pt} {\kern 1pt} {\kern 1pt} m{h}\ifmmode{'}\else$'$\fi_{1} - {C}\ifmmode{'}\else$'$\fi_{1} m{h}\ifmmode{'}\else$'$\fi_{2} } \right)}} } > 0. $$

Recall that h2 < 0 if the income effect on the demand for real balances outweighs the wealth effect. Then, the same analysis as in the former case leads to the following equation that determines the slope of the stable arm in the saddle-point equilibrium:

$$ {\text{NFA}} - {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}} = - \frac{{\mu {\kern 1pt} {\kern 1pt} _{1} + m{h}\ifmmode{'}\else$'$\fi_{1} }} {{m{h}\ifmmode{'}\else$'$\fi_{2} }}{\left( {m - \ifmmode\expandafter\bar\else\expandafter\=\fi{m}} \right)}. $$
(A7)

Therefore, the stable arm is downward sloping for the case in which changes in NFA are produced by changes in foreign liabilities only.

Appendix 3 Comparative static analysis of macroeconomic policies

Setting (2) and (5) to zero and totally differentiating them with respect to \(\overline{{{\text{NFA}}}} ,\,\ifmmode\expandafter\bar\else\expandafter\=\fi{m}\), g M , and G produces

$$ {\left( {\begin{array}{*{20}c} {{ - {C}\ifmmode{'}\else$'$\fi_{1} }} & {{ - {C}\ifmmode{'}\else$'$\fi_{1} }} \\ {{ - m{h}\ifmmode{'}\else$'$\fi_{2} }} & {{ - m{h}\ifmmode{'}\else$'$\fi_{1} }} \\ \end{array} } \right)}{\left( {\begin{array}{*{20}c} {{d{\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} \\ {{d\ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} \\ \end{array} } \right)} = {\left( {\begin{array}{*{20}c} {{dG}} \\ {{ - mdg_{M} }} \\ \end{array} } \right)}.\, $$
(A8)

Applying the Cramer’s rule yields

$$ \frac{{\partial {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} {{\partial g_{M} }} = \frac{1} {{{h}\ifmmode{'}\else$'$\fi_{2} - {h}\ifmmode{'}\else$'$\fi_{1} }} > 0,\,\frac{{\partial \ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} {{\partial g_{M} }} = \frac{1} {{{h}\ifmmode{'}\else$'$\fi_{1} - {h}\ifmmode{'}\else$'$\fi_{2} }} < 0; $$
(A9)
$$ \frac{{\partial {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} {{\partial G}} = - \frac{{{h}\ifmmode{'}\else$'$\fi_{1} }} {{{C}\ifmmode{'}\else$'$\fi_{1} {\left( {{h}\ifmmode{'}\else$'$\fi_{1} - {h}\ifmmode{'}\else$'$\fi_{2} } \right)}}} < 0,\,\frac{{\partial \ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} {{\partial G}} = \frac{{{h}\ifmmode{'}\else$'$\fi_{2} }} {{{C}\ifmmode{'}\else$'$\fi_{1} {\left( {{h}\ifmmode{'}\else$'$\fi_{1} - {h}\ifmmode{'}\else$'$\fi_{2} } \right)}}} < 0. $$
(A10)

Similarly, following the same analytical procedure for the case in which foreign liabilities rather than foreign assets drive changes in NFA yields

$$ {\left( {\begin{array}{*{20}c} {{ - {F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{2} } \right)} - {C}\ifmmode{'}\else$'$\fi_{1} }} & {{ - {C}\ifmmode{'}\else$'$\fi_{1} }} \\ {{ - m{h}\ifmmode{'}\else$'$\fi_{2} }} & {{ - m{h}\ifmmode{'}\else$'$\fi_{1} }} \\ \end{array} } \right)}{\left( {\begin{array}{*{20}c} {{d{\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} \\ {{d\ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} \\ \end{array} } \right)} = {\left( {\begin{array}{*{20}c} {{dG}} \\ {{ - mdg_{M} }} \\ \end{array} } \right)}. $$
(A11)

Applying the Cramer’s rule produces

$$ \frac{{\partial {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} {{\partial g_{M} }} = - \frac{{m{C}\ifmmode{'}\else$'$\fi_{1} }} {{{\left| {J_{2} } \right|}}} > 0,\,\frac{{\partial \ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} {{\partial g_{M} }} = \frac{{m{\left[ {{F}\ifmmode{'}\else$'$\fi{\left( {1 - {C}\ifmmode{'}\else$'$\fi_{2} } \right)} + {C}\ifmmode{'}\else$'$\fi_{1} } \right]}}} {{{\left| {J_{2} } \right|}}} < 0; $$
(A12)
$$ \frac{{\partial {\text{\ifmmode\expandafter\bar\else\expandafter\=\fi{N}\ifmmode\expandafter\bar\else\expandafter\=\fi{F}\ifmmode\expandafter\bar\else\expandafter\=\fi{A}}}}} {{\partial G}} = - \frac{{m{h}\ifmmode{'}\else$'$\fi_{1} }} {{{\left| {J_{2} } \right|}}} < 0,\,\frac{{\partial \ifmmode\expandafter\bar\else\expandafter\=\fi{m}}} {{\partial G}} = \frac{{m{h}\ifmmode{'}\else$'$\fi_{2} }} {{{\left| {J_{2} } \right|}}} > 0, $$
(A13)

where ∣J 2∣ is defined as in Eq. A3.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Wu, Y. A macroeconomic analysis of foreign assets and foreign liabilities. J Econ Finan 32, 119–135 (2008). https://doi.org/10.1007/s12197-007-9014-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12197-007-9014-2

Keywords

JEL Classifications

Navigation