Abstract
In this paper, we analyze a heterogeneous agent model in which the fundamental exchange rate is endogenously determined by the real markets. The exchange rate market and the real markets are linked through the balance of payments. We have analytically found that there exists at least a steady state in which the exchange rate is equal to its fundamental value and incomes of both countries are equal to the autonomous components times the multiplier (as in the IncomeExpenditure model). This steady state can be unique and unstable when all agents act as contrarians, while when agents act as fundamentalists it is unique but its stability depends on the reactivity of actors of the market. Finally, we show that the (in)stability of the economic system depends on both the reactivity of the markets and that of different types of agents involved. Employing wellknow functional forms, we show that the model can replicate some of the statistical features of the true time series of the exchange rate.
Similar content being viewed by others
Notes
Our model is close to that developed by Laursen and Metzler in 1950. The Laursen and Metzler model (1950) is static, while its continuous time dynamic version can be found in Gandolfo (1986).
As stated in De Arcangelis and Gandolfo (1997), we do not adhere to the traditional approach to the exchange rate, where the determination of the exchange rate is ’reflected in the balance of payments equation’, which summarizes the foreign exchange market. Moreover, our formulation implies a unitary elasticity.
In line with empirical works, this specification implies that the exchange rate depends only on speculators and not on import/export exchanges.
This is a shortmediumrun model so that we believe that the assumption of fixed prices is not so stringent.
Obviously the demand can absorb all the product. However, when there is an excess of demand, we assume that both countries have to ration all components of the demand other than export.
This is the only steady state having that property.
References
Anufriev M (2013) Heterogeneous Beliefs in an asset pricing model with endogenous fundamentals. Mimeo
Brock W, Hommes C (1997) A rational route to randomness. Econometrica 65:1059–1095
Brock W A, Hommes C H (1998) Heterogeneous beliefs and routes to chaos in a simple asset pricing model. J Econ Dyn Control 22:1235–74
Day R, Huang W (1990) Bulls, bears and market sheep. J Econ Behav Organ 14:299–329
De Arcangelis G, Gandolfo G (1997) Nonlinear phenomena in economics: the exchange rate. Nonlinear Anal Theory Appl 30:1043–1049
De Grauwe P, Rovira P K (2012) Animal Spirits in the foreign exchange market. J Econ Dyn Control 36:1176–1192
Federici D, Gandolfo G (2012) The Euro/Dollar exchange rate: Chaotic or nonchaotic? A continuous time model with heterogeneous beliefs. J Econ Dyn Control 36:670–681
Gandolfo G (1986) Economia Internazionale, vol II, Ed. UTET, Torino
Hommes C H (2006) Heterogeneous agent models in economics and finance. In: Tesfatsion L, Judd K L (eds) Handbook of computational economics, volume 2: agentbased computational economics. Elsevier, pp 1109–1186
Kirman A (1998) On the transitory nature of gurus. Mimeo
Laursen S, Metzler L (1950) Flexible exchange rate and the theory of employment. Rev Econ Stat 32: 281–99
LeBaron B (2006) Agentbased computational finance. In: Tesfatsion L, Judd K L (eds) Handbook of computational economics, volume 2: agentbased computational economics. Elsevier, pp 1187–1233
Lengnick M, Wohltmann H W (2013) Agentbased financial markets and new keynesian macroeconomics  a synthesis. J Econ Interac Coord 8:1–32
Naimzada A K, Ricchiuti G (2008) Heterogeneous Fundamentalists and Imitative Processes. Appl Math Comput 199:171–180
Naimzada A K, Ricchiuti G (2009) Dynamic Effects of Increasing Heterogeneity in Financial Markets. Chaos, Solitons & Fractals 41:1764–1772
Naimzada A K, Pireddu M (2014) Dynamic Behavior of Product and Stock Markets with a Varying Degree of Interaction. Econ Model 41:191–197
Proaño C R (2011) Exchange Rate Determination, Macroeconomic Dynamics and Stability under Heterogenous Behavioural FX expectations. J Econ Behav Organ 77:177–188
Rovira P K (2010) Uncertainty about fundamentals and herding behavior in the FOREX market. Physica A 389:1215–1222
Westerhoff F (2009) Exchange rate dynamics: a nonlinear survey. In: Rosser Jr, J B (ed) Handbook of Research on Complexity. Edward Elgar, Cheltenham, pp 287–325
Westerhoff F (2012) Interactions between the real economy and the stock market: A simple agentbased approach. Discrete Dynamics in Nature and Society
Acknowledgements
We really thank an anonymous referee and the Editor, Uwe Cantner, for suggestions that improved the paper. We also wish to thank Marco Dardi and Frank Westerhoff who discussed with us an early version of the paper, the participants in the workshops held in Scuola Normale di Pisa in December 2013 and in Univerisità MilanoBicocca in February 2014, and the participants in the conference Computing in Economics and Finance (CEF 2014) held in Oslo in June 2014.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interests
The authors declare that they have no conflict of interest.
Appendix: Proof of Proposition 1
Appendix: Proof of Proposition 1
Defining y _{1}(t) = m p i _{1} Y _{1}(t) and y _{2}(t) = m p i _{2} Y _{2}(t), we can write system (9) as follows
Defining now
we get the following system
Denoting by \(\widehat {S}\) the set of steady states of Eq. 15, we have that the function \({\Phi }: \widehat {S}\to S\) mapping \((y_{1},y_{2},E)\in \widehat {S}\) into (y _{1}/m p i _{1},y _{2}/m p i _{2},E) ∈ S is well defined and is a bijection. Moreover, \((y_{1},y_{2},E)\in \widehat {S}\) is asymptotically stable if and only if Φ(y _{1},y _{2},E) ∈ S is asymptotically stable, and \((y_{1},y_{2},E)\in \widehat {S}\) is unstable if and only if Φ(y _{1},y _{2},E) ∈ S is unstable. As a consequence, we can prove Proposition 1 about system (9) working with system (15) and properly using the bijection Φ and equalities (14).
Note first that, for every \(s\in \mathbb {R}_{++}\),
Since the system
has a set of solutions given by
we have that
where Γ is defined in Eq. 10, and that immediately implies the first statement of Proposition 1. Note also that
Then
belongs to \(\widehat {S}\) and since \({\Phi }(y^{*}_{1},y^{*}_{2},E^{*})\) is equal to Eq. 11 and belongs to S, we get the second statement of Proposition 1.
Assume now that, for every \((z_{1},z_{2})\in \mathbb {R}_{++}^{2}\), we have that
and prove that \(\widehat {S}=\{(y^{*}_{1},y^{*}_{2},E^{*})\}\), so that \(S=\{\Phi (y^{*}_{1},y^{*}_{2},E^{*})\}\). Indeed,
if and only if
that is,
and that equation has E ^{∗} has unique solution. That proves the third part of Proposition 1.
Let us move on to prove the last part of the proposition. We analyze then the stability of the state \((y_{1}^{*},y_{2}^{*},E^{*})\) defined in Eq. 17 in order to deduce information about the stability of \({\Phi }(y^{*}_{1},y^{*}_{2},E^{*})\), that is, Eq. 11. Consider the three following functions the domain of which is \(\mathbb {R}_{++}^{3}\),
Then we have that:
Note that
where R is defined in Eq. 12, and, for every j ∈{1,…,n} and \(s\in \mathbb {R}\),
Then we have
In order to study the stability of Eq. 17, we study the stability of the polynomial
that is, we study whether all of its roots lie inside the unit circle or there is at least one root lying outside the unit circle. A computation shows that
where
A simple condition which is sufficient for instability is P(1) < 0, that is, ρ _{3} < 0. Since each factor in \(f_{1}^{\prime }(0)f^{\prime }_{2}(0)(b_{1}b_{2}b_{1}b_{2}+1)\) is positive, we have that P(1) < 0 if and only if R < 0. That proves Statement A1 of Proposition 1.
Another simple condition for instability is P(−1) > 0, that is, − 8 + 4ρ _{1} − 2ρ _{2} + ρ _{3} > 0. Note that,
if and only if
Since
and \(f^{\prime }_{1}(0),f^{\prime }_{2}(0)>0\), we have that Eq. 18 is implied by
that is
It can be verified that Eq. 19 holds true if one of the following set of conditions is satisfied:

\(\frac {2}{b_{1}}<f_{1}^{\prime }(0)<\frac {2}{b_{1}1}\), \(f_{2}^{\prime }(0)<\frac {2}{b_{2}}\), and R > 0;

\(f_{1}^{\prime }(0)<\frac {2}{b_{1}}\), \(\frac {2}{b_{2}}<f_{2}^{\prime }(0)<\frac {2}{b_{2}1}\), and R > 0;

\(f_{1}^{\prime }(0)>\frac {4}{b_{1}1}\), \(f_{2}^{\prime }(0)>\frac {4}{b_{2}1}\), \(R>2\frac {(b_{1}+1)(b_{2}+1)}{(b_{1}1)(b_{2}1)}\).
Those conditions are exactly the ones in the Statements A2, A3 and A4 of Proposition 1.
In order to prove the last part of the proposition, assume m p c _{1} = m p c _{2}, m p i _{1} = m p i _{2}, \(g^{\prime }_{1}(0)=g^{\prime }_{2}(0)\), and R > 0. That implies b _{1} = b _{2} and \(f_{1}^{\prime }(0)=f_{2}^{\prime }(0)\) and
where we set b = b _{1} and \(f=f^{\prime }_{1}(0)\). Let P(s) = (s − 1)K(s) + R J(s), where
and
The equation (s − 1)K(s) = 0 has three distinct solutions given by
with s _{1} < s _{2} < 1, and since J(s) = [(s − 1) + f(b − 1)]^{2} = (s − s _{2})^{2}, we also have that P(s _{2}) = 0, P ^{′}(s _{2}) < 0, P(s _{1}) > 0 and, for every s ∈ [1, + ∞), P(s) > 0. As a consequence P has three distinct real roots \(s^{*}_{1}<s^{*}_{2}<s^{*}_{3}\) such that
As a consequence, if \(f\ge \frac {2}{b+1}\), then \(s^{*}_{1}<s_{1}\le 1\) and Eq. 17 is unstable: that proves Statement B1 of Proposition 1.
Assuming instead \(f<\frac {2}{b+1}\), we get s _{1} > −1 and then, \(s^{*}_{1}<1\) if and only if P(−1) > 0, while \(s^{*}_{1}\in (1,1)\) if and only if P(−1) < 0.
A computation shows that
and then P(−1) > 0 if and only if
if and only if
Analogously, we have P(−1) < 0 if and only if
In the first case we get instability while in the second case we get asymptotic stability: by a substitution we obtain conditions of Statements B2 and B3 of Proposition 1.
Rights and permissions
About this article
Cite this article
Gori, M., Ricchiuti, G. A dynamic exchange rate model with heterogeneous agents. J Evol Econ 28, 399–415 (2018). https://doi.org/10.1007/s0019101705139
Published:
Issue Date:
DOI: https://doi.org/10.1007/s0019101705139