ArticlePDF Available

Crises and capital controls in small open economies: a stock–flow consistent approach

Authors:

Abstract and Figures

This paper attempts to explain the role of capital inflows in creating economic booms and busts in a small open economy with sovereign currency. We develop a stock–flow consistent (SFC) model for a small open economy while relying on the experience of the Icelandic crisis. We demonstrate the destabilising effects of capital inflows on the economy by allowing for a sudden stop, and also discuss the role of capital controls as a policy response in the event of a crisis due to sudden stops. Finally, we discuss the policy implications of our results in order to tackle the destabilising effects associated with financial flows in a small economy.
Content may be subject to copyright.
Crises and capital controls in small open
economies: a stockflow consistent approach*
Hamid Raza**
Aalborg University, Denmark
Bjorn Runar Gudmundsson***
Statistics Iceland, Reykjavik, Iceland
Gylfi Zoega****
University of Iceland, Reykjavik, Iceland
Mikael Randrup Byrialsen*****
Aalborg University, Denmark
This paper attempts to explain the role of capital inflows in creating economic booms and busts in a
small open economy with sovereign currency. We develop a stockflow consistent (SFC) model for a
small open economy while relying on the experience of the Icelandic crisis. We demonstrate the desta-
bilising effects of capital inflows on the economy by allowing for a sudden stop, and also discuss the
role of capital controls as a policy response in the event of a crisis due to sudden stops. Finally, we
discuss the policy implications of our results in order to tackle the destabilising effects associated with
financial flows in a small economy.
Keywords: post-Keynesian, stockflow consistent, monetary policy, capital controls, sudden stops,
financial crisis
JEL codes: E12, F32, F38, F41
1 INTRODUCTION
Lifting restrictions on capital movements has proven to be a double-edged sword for
many countries as expectations of greater prosperity have given way to financial turmoil
and crisis. Iceland is a prime example where a short period of open capital markets gave
rise to intensive domestic growth fuelled by exuberant capital inflows. This eventually
turned into a nightmare scenario when access to capital markets dried up, dragging
the economy into a recession in late 2008.
* This work was supported by the Icelandic Research Fund (IRF grant number 130551-051).
** Email: raza@business.aau.dk.
*** Email: Bjorn.Gudmundsson@hagstofa.is.
**** Email: gz@hi.is.
***** Email: randrup@business.aau.dk.
Received 21 February 2018, accepted 28 July 2018
European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1, 2019, pp. 94133
First published online: February 2019; doi: 10.4337/ejeep.2019.0042
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
The Lypiatts, 15 Lansdown Road, Cheltenham, Glos GL50 2JA, UK
and The William Pratt House, 9 Dewey Court, Northampton MA 01060-3815, USA
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
While the Icelandic story is relatively well-known in general,
1
few attempts have been
made to explicitly model the origins and channels of capital inflows to Iceland, which led
to the build-up of external imbalances. A significant proportion of the capital inflows can
be traced to bond issuance by the Icelandic banking sector held by international portfolio
investors. This made fixed income securities one of the main sources of credit inflow in
the first years of financial expansion. Another main source of the inflows into Iceland was
the retail deposit accounts offered by the Icelandic banks during the last couple of years
prior to the financial crisis. By offering significantly higher deposit rates, Icelandic banks
were able to collect huge amounts of deposits, allowing them to reduce their capital mar-
ket exposure while maintaining high growth momentum.
Capital inflows can induce financial sector growth, which in turn can have strong
impacts on the real sector of the economy. Thus, explicitly modelling capital inflows
alongside a financial and real sector enables us to understand the instability associated
with capital inflows. The role of capital inflows in particular is crucial for understanding
the balance of payments and exchange-rate dynamics in a small open economy with a
sovereign currency such as Iceland.
Traditional models of exchange-rate determination tend to play down the role of capi-
tal flows, treating them as an independent source of exchange-rate movements. The main
focus of these models has been centred on the trade flows, usually within an inter-
temporal optimising set-up.
2
In such settings, movements on the capital account play a
passive role relative to the current account while resting on the usual assumptions of pur-
chasing-power parity and uncovered-interest parity. The Icelandic experience is somewhat
at odds with the above view. Iceland was able to maintain a current-account deficit for a
long time, suggesting that access to foreign financing was easily forthcoming despite
obvious macroeconomic imbalances.
With the rise in cross-border flows, many models acknowledged the role of capital mar-
kets in explaining the balance-of-payments and exchange-rate dynamics. Most of these
models, however, considered the balance of payments in isolation
3
or ignored the special
role played by the banks in attracting financial flows
4
(Benes et al. 2014). Thus, the effects
of capital inflows on the real economy remains largely unclear. Similarly, the role of capital
controls, as described by Eichengreen (2004), is one of the most controversial and least
understood issues in macroeconomics.
This paper attempts to explain the role of capital inflows in creating economic booms
in a small open economy with sovereign currency. We develop a stockflow consistent
(SFC) model for a small open economy while relying on the experience of the Icelandic
crisis. We demonstrate the destabilising effects of capital inflows on the economy by
allowing for a sudden stop, and also discuss the role of capital controls in response to a
crisis due to sudden stops.
The paper is organised as follows. Section 2 briefly discusses the history of economic
instabilities occurring due to cross-border flows. Section 3 presents the structure of the
model for a very small open economy with a floating currency, and also explains the results
while introducing different regimes of capital flows. Section 4 concludes the paper.
1. For example, Matthiasson (2008); Benediktsdottir et al. (2011); Baldursson/Portes (2013);
Gudmundsson (2016); Raza et al. (2016).
2. See Dornbusch (1975) for an explanation.
3. For example, Magud et al. (2011).
4. For example, Engel/Matsumoto (2006). While some progress has been made since the crisis,
Benes et al. (2014) argue that substantial work is still required in order to examine the role of the
financial sector from a balance-sheet perspective in particular.
Crises and capital controls in small open economies: a stockflow consistent approach 95
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
2 FINANCIAL FLOWS AND INSTABILITIES
Since the late 1980s, many countries in pursuit of financial development abolished capital
controls and refrained from interventions in the financial markets. Financial-account
openness in an environment of innovative financial markets during this period greatly
facilitated access to international capital. Less emphasis was placed on the risks associated
with foreign borrowing. This allowed a sizeable expansion of the financial sector relative to
the real sector in small economies, reaching levels never experienced before in the available
statistics.
5
This resulted in an extremely heavy reliance on finance-led growth. The expan-
sion of the financial sector took place regardless of the prevailing currency regimes, for
example, European peripheral countries (currency union), the Baltics (fixed exchange
rate) and Iceland (floating regime).
With the eruption of the global financial crisis (GFC), small open economies that were
heavily reliant on international credit experienced capital reversals, balance-of-payments crisis
and recessions, all resulting in high unemployment, large fiscal costs and spending cuts on
welfare. The crisis in 20072008 revealed two fundamental but interrelated weaknesses in
small open economies. First, it revealed the lack of resilience of these economies to with-
stand the effects of global shocks. Second, it exposed the destabilising effects of financial
flows in their systems, which had largely been ignored before the crisis. The latter can per-
haps be linked to the psychology of policy-makers explained in Reinhart/Reinhart (2008),
that persistent inflows can lead to the belief that the episode is a permanent phenomenon
rather than a temporary one.
The surge in cross-border financial flows linked to economic integration and the sub-
sequent decline generating economic instabilities is not unique to the recent episode of
20072008. The asset boom fuelled by capital inflows can be traced back to the Railway
Mania in Britain in the 1840s, which ended with a banking crisis and a severe economic
recession in 1847 (Bordo/Landon-Lane 2013). While past experiences of booms and busts
share many similarities, they also have vital differences. These differences provide impor-
tant lessons for policy-makers to design better policies and regulations.
In the past 140 years, the surges in global capital flows can broadly be divided into five
major episodes.
6
The first episode, 18701913, was characterised by a boom in bond financing from
advanced economies to the regions of recent settlement with abundant resources
and scarce labour. The investors preferred fixed income debt instruments while for-
eign direct investment (FDI) share was small.
The second episode, from post-World War I to the Great Depression, was marked
by a large investment in government bonds. The governments heavily borrowed to
finance public expenditures.
7
The third episode, 19731981, was driven by oil price shocks. The flows, however,
were dominated by bank lending to finance the balance of payments in developing
countries, while the share of debt instruments remained low.
5. The relationship between capital inflows and domestic credit growth is discussed in Lane/
McQuade (2014).
6. See Bordo/Landon-Lane (2013) for a discussion on the historical perspective of asset prices,
booms and busts. Also see Accominotti/Eichengreen (2016) on the history of capital inflows and
reversals in Europe.
7. During these two episodes, credit was volatile but remained stable relative to the size of the
economy in the long run. The only exception was the Great Depression of the 1930s, during
which credit relative to the size of the economy collapsed (Schularick/Taylor 2009).
96 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
The fourth episode, 19931997, marked the surge in capital flows to the private
borrowers in emerging markets. The composition of capital flows in this episode
greatly changed with the rise in FDI share along with investments in bonds and
equities, while the share of bank lending was low. On the borrowing end, the
share of private-sector debt dramatically increased.
Fifth, the recent episode of the 2000s, with increased short-term inflows. Banks had
cheap access to international credit, which amongst other things resulted in unpre-
cedented levels of lending to the private sector in general.
In all major episodes of capital booms, real output growth was accompanied by asset
price booms and strong growth in international investment and trade. All episodes were
eventually followed by a bust, causing currency crisis, financial instabilities and real output
losses in the capital-borrowing economies. The impact of the crisis on the real economy in
2008 was the most severe since the Great Depression. The average output loss as a result
of the recent crisis, reported by Laeven/Valencia (2010), is 25 per cent as compared to a
historical average of 20 per cent estimated for past crises. In most country-specific cases,
the underlying factor triggering the crisis has been an abrupt stop of inflows, beyond the
control of the recipient economy.
8
Despite a large literature on the economic volatility associated with inflows, restrictions
on capital inflows are seen with scepticism. Ghosh/Qureshi (2016) discuss the long history
of capital controls and explain how thinking around them has evolved over time. An inter-
esting explanation of scepticism towards capital controls put forward by the authors is that
people usually think that controls on inflows and outflows are linked, whereas the latter
have been typically associated with autocratic and repressive regimes. The authors argue
that controls on inflows as an economic policy to reduce volatility has not been specifically
analysed from a historical perspective.
After the recent Great Recession, large financial flows relative to the size of the econ-
omy and their effects on growth and stability have become a topic of debate in macro-
economics.
9
Small economies benefited from globalisation and the resultant
international capital flows, but this also posed a serious challenge to the policy-makers,
essentially due to loss of monetary control, loss of competitiveness and vulnerability to
external shocks.
10
There seems to be an emerging consensus that a very large financial
sector amplified by large financial inflows is potentially destabilising. For instance,
Reinhart/Reinhart (2008) find that capital inflow bonanzas are associated with a higher
likelihood of economic crisis. The emerging literature following the financial crisis sug-
gests that capital controls may be justified when capital flows are strong and the econ-
omy fragile. In a recent paper published in the IMF Economic Review, Farhi/Werning
(2014) find that in a New Keynesian model capital controls are desirable even when the
exchange rate is flexible. Korinek/Sandri (2016) argue that capital controls are desirable in
emerging economies because they stimulate domestic saving by favouring domestic interme-
diation. Zeev (2017) studies 33 emerging market economies and finds that output in econo-
mies with stricter capital inflow controls responds less to global credit supply shocks while
capital outflow controls have no significant shock-absorbing capacity. A recent book by
8. The only exceptions highlighted in the literature are Chile (19901991) and Malaysia
(19931994), where sudden stops were preceded by voluntary implementation of restrictions
on short-term inflows (Calvo/Reinhart 1999).
9. See, for example, Obstfeld (2012); Krugman (2014).
10. The challenges posed to policy-makers by globalisation are well described in the political tri-
lemma of world economy by Rodrik (2000).
Crises and capital controls in small open economies: a stockflow consistent approach 97
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Ghosh et al. (2017) studies the effectiveness of capital controls in a large number of coun-
tries, the tools used to implement capital controls and their pros and cons.
The experience of Iceland regarding capital inflows is somewhat similar to other crisis-hit
economies, but nonetheless provides an excellent testing ground due to its exceptional
financial sector growth, followed by a regime of strong capital controls, leading to a different
post-crisis scenario than other countries.
11
Using the Icelandic experience as a motivation,
we now proceed to derive a theoretical model of capital inflows in the SFC tradition.
3 MODELLING CAPITAL INFLOWS IN A VERY SMALL OPEN ECONOMY
In general, modelling an open economy is a tough challenge due to cross-border capital flows
originating from complex financial systems. However, the balance-sheet approach in the SFC
framework conveniently allows one to incorporate the dynamics of international capital inflows
by explicitly modelling the portfoliochoices of domestic and international traders. In addition,
SFC models are now well-known for their ability to link the real and financial sectors of the
economy.
12
In the context of the recent financial crisis in Iceland, these features are central
to our study. Therefore, the SFC approach provides a solid platform for studying the crisis.
Several authors have attempted to extend SFC analysis from closed to open economies.
A very basic framework of an open economy with no cross-border financial movements is
discussed in chapter 6 of Godley/Lavoie (2007). The analysis is then extended to a more
advanced model in chapter 12 by introducing a number of realistic features such as flexible
exchange rates, trading foreign assets and settling official imbalances through foreign currency
denominated assets. Lavoie/Daigle (2011) further extended chapter 12 of Godley/Lavoie
(2007) while mainly focusing on the issue of exchange-rate expectations. Duwicquet/Mazier
(2010) developed a model for two countries of asymmetric size in a monetary union. They
introduced three different shocks (loss of competitiveness, fall in consumption and a reduc-
tion in capital accumulation) to analyse macroeconomic adjustment and stabilisation in a
monetary union. Kinsella/Khalil (2012) used a two-country SFC model to study a small
economy experiencing debt deflation. They performed a series of simulations, including
the impact of investment shock on debt deflation in a small economy under two different
currency regimes (floating exchange rate and currency union). Greenwood-Nimmo (2014)
examined the role of fiscal and monetary policy in an open-economy SFC model. A recent
study by Burgess et al. (2016) developed an open-economy SFC model for the UK economy
to address financial balances.
Open-economy models in the SFC tradition emphasise the importance of complete-
ness and reciprocity in order to catch the dynamics of economic interconnections of
countries trading with each other (see Lavoie/Daigle 2011, Mazier/Tiou-Tagba Aliti
2012, Greenwood-Nimmo 2014 and Caiani et al. 2018, amongst others). However,
studies considering small open economies relax these conditions and treat developments
within the rest of the world as given.
13
In these models, the small open economy is trea-
ted as the one influenced by global factors, but not vice versa. A few recent examples of
these are Meijers et al. (2015) and Byrialsen/Raza (2018).
11. Raza et al. (2018) empirically investigate the current-account adjustment in Iceland and com-
pare it to that of Ireland and other eurozone countries.
12. For a comprehensive survey on the SFC approach to modelling, see Caverzasi/Godin (2015)
and Nikiforos/Zezza (2017).
13. Note that this still involves modelling trade flows and portfolio allocation of international
traders.
98 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Built around the same general idea, our model also adopts the small country assumption
for Iceland, assuming the negligible impact of the Icelandic economy on the rest of the
word. As a consequence, only model equations for the domestic economy need to be con-
sidered. Before explaining the structure of the model, we explain the balance sheets and
transaction flows in our model.
3.1 Transaction flow and balance-sheet matrices
First, we present the interactions between the balance sheets of different sectors in the
economy. A formal representation of assets and liabilities can be seen in the balance-
sheet matrix in Table 1. The items reported with plus (+) signs represent assets, while
minus () signs represent liabilities. As can be seen from the balance-sheet matrix, finan-
cial stocks in our model take the form of deposits, bills and loans. The banking sector has
two branches: (i) a domestic branch, which holds domestic assets on its balance sheets; and
(ii) a foreign branch, which attracts foreign deposit holders.
We now proceed to an explanation of the transactions in the economy. A complete
presentation of the transactions flows matrix can be seen in Table A1 in Appendix 1.
The matrix tracks all the flows between different sectors within the economy as well as
the cross-border flows. A plus (+) sign in the matrix indicates that the flow has been
received or that it can simply be considered as an income, while a minus () sign repre-
sents an outflow or simply an expenditure. For simplicity, aggregated transaction flows are
graphically presented in Figure 1. There are three types of transactions taking place in the
economy: (i) transactions associated with real economic activity; (ii) transactions asso-
ciated with income from capital, which are interest rates in this case; and (iii) transactions
occurring due to changes in financial assets and liabilities.
14
The production in the economy takes place as described in the standard GDP identity.
Trade and production of goods take place in the firm sector. Firms pay the wage bill to
their workers (household) and taxes to the government. They acquire loans, which cover
the discrepancy between investment expenditures and profits. They also pay interest on
their loans to the financial sector.
Households finance their consumption by receiving income in the form of wages and
interest payments on deposits and bills. They pay income taxes to the government sector.
The government sector receives income in the form of taxes from the firms and the house-
holds. It adjusts its expenditures according to its revenue, hence running a balanced budget.
Table 1 The balance-sheet matrix
Small economy RoW Net
Household Firms Banks Foreign branch RoW
Deposits þDice
DbþDrow 0
Loans
LfþLb0
Fixed capital þK––
þK
Banks bills þBice
BbþBrow 0
RoW bills þBrow
ice
––
Brow 0
14. Note that the change in financial stock in the graph refers to the aggregation of financial assets
and liabilities presented in Table 1.
Crises and capital controls in small open economies: a stockflow consistent approach 99
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
The banking sector as a whole offers both the domestic currency denominated loans
and the foreign currency denominated loans. The banking sector receives income in
the form of interest on its lending. Banks also issue bills internationally to meet the
demand for credit in the economy. They pay interest on their liabilities, which are deposits
and bills. Finally, the rest of the world receives interest from Iceland on their bills and
deposits, and engages in trade with Iceland.
In our model, the balance-of-payments dynamics are driven by the negative net posi-
tion of the corporate sector. This setting is consistent with the sectoral balances of Iceland
before the crisis. In Figure 2, it can be seen how the position of Iceland relative to the
RoW sector (which is the balance-of-payments deficits and accumulation of external
debt) is driven by the negative net position of the corporate sector. While the banks rela-
tively maintain a net position somewhat closer to zero, they fulfil (as well as induce) the
demand for credit in the economy by holding external debt on their balance sheets, which
makes them vulnerable to a sudden stop.
3.2 Structure of the model
The structure of our model makes clear reference to the exceptional growth of the Icelandic
financial sector. The institutional elements of the model relate to the specific social and eco-
nomic conditions that evolved during the early 2000s, when Icelandic banks could attract
huge amounts of portfolio capital seemingly without raising concerns about systemic sus-
tainability. By implementing policies such as financial sector privatisation and deregulation
of market activity, Iceland was seen as following a benign recipe for institutional
Figure 1 Transaction flows
100 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
improvement. Obviously some concerns were raised, and during the mini-crisis of early
2006, access to financial markets was temporarily restricted. However, confidence was
restored as markets became convinced once again of the sustainability of the Icelandic finan-
cial model following protective measures by the Icelandic banks and assurances by the
authorities that public support would be forthcoming if needed.
With the GFC, it became gradually more and more evident that the likelihood of systemic
failurecouldnotbeexcludedandinvestorsconfidence disappeared, resulting in a sudden stop.
This fits well into the Harvey (2009) framework, explaining the role of psychological or beha-
vioural factors, with regard to exchange-rate dynamics. The role of market psychology was
perhaps even more important during the earlier stages when financial expansion took place
in the early 2000s. The newly privatised Icelandic banks rushed into expansion mode, attract-
ing large amounts of capital supporting further expansion. Bandwagon effects or Chartist
investment strategies clearly seemed to be driving factors while economic fundamentals
such as current-account imbalances were given less attention until the eruption of the crisis.
In order to capture the above aspects, the structure of the model is specified as shown
below.
15
3.2.1 Firms
Firms produce goods by employing workers ðNÞwhile paying them wages ðWÞ. Prices Ps
are set as a mark-up ðϕÞover unit cost (equations (1)(3)).
National income
y¼cþgþiþxm(1)
−30
−20
−10
0
10
20
Percentage of GDP
2000 2005 2010 2015
Non-financial corporations
Rest of the world
Financial corporations
Household sector Government sector
Figure 2 Sectoral balances of the Icelandic economy
15. Note that small letters indicate real (deflated) variables whereas capital letters represent nom-
inal values. The full model can be seen in Appendix 1.
Crises and capital controls in small open economies: a stockflow consistent approach 101
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Wage bill
WB ¼WðNÞ(2)
Ps¼ð1þUCÞϕ(3)
Following the partial adjustment accelerator model, the level of real investment (i)inthe
model is determined by a partial adjustment of the existing stock of capital towards a
desired stock of capital kT, plus a level of depreciation (da) (equations (4)(7)).
Real investment
i¼γðkTkt1Þþda (4)
Targeted stock of capital
kT¼ηðyt1Þ(5)
Depreciation of capital
da ¼δðkt1Þ(6)
Change in stock of capital
Δk¼ida (7)
Profit of firms
Ff¼YWB Tf(8)
Profits ðFfÞafter paying taxes ðTfÞare used to finance investment in every period. If
the targeted investment is higher than the profits, which is the case in our model, investment
is financed by directly borrowing from the banks (equation (9)). Firms can borrow both the
domestic currency denominated loans ðLf;ice;dÞand foreign currency denominated loans
ðLf;row;dÞasshowninequations(10)(11). This feature is consistent with the experience of
Icelandic firms, which borrowed both domestic and foreign currency denominated loans. In
Icelands particular case, two main motives for foreign-exchange (FX) denominated loans
can be identified as follows: (i) a proportion of firmsrevenue coming from exports was
in the foreign currency; and (ii) interest rates charged on domestic denominated loans was
higher than on the foreign currency denominated loans. Based on this argument, the demand
for loans in our model has an inverse relationship with the associated interest rate, that is, if
the interest rate on domestic currency borrowing ðrL;iceÞincreases as compared to the interest
rate on FX denominated loans ðrL;rowÞ, the proportion of borrowing in the domestic currency
will fall while the proportion of borrowing in FX denominated loans will increase.
Demand for loans
Łf;d¼Lf;d
t1þIFfþrL;ice
t1ðLf;ice;d
t1ÞþrL;row
t1ðLf;row;d
t1Þ(9)
Demand for Icelandic króna (ISK) and foreign denominated loans
16
Lf;ice;d¼Lf;dðψ1ψ2rL;ice
ice þψ3rL;row
ice Þ(10)
Lf;row;d¼Lf;dðψ4þψ5rL;ice
ice ψ6rL;row
ice Þ(11)
16. Note that the sum of the demand for domestic and FX loans equals total loans, that is,
Łf;d¼Lf;ice;dþLf;row;d.
102 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
3.2.2 Households
Householdsconsumption ðcÞis a function of disposable income ðydhsÞadjusted
for exchange-rate gains on foreign assets and their past wealth, ðvÞas can be seen
in equations (12)(15).
Consumption function
c¼α1ðydhsÞþα2ðvt1Þ(12)
Householdsdisposable income
YD ¼WB þrB
iceðt1ÞðBice;d
iceðt1ÞÞþrB
rowðt1ÞðBrow;d
iceðt1ÞÞþrD
iceðt1ÞðDh;ice;d
iceðt1ÞÞTh(13)
HaigSimons disposable income
YDHS ¼YD þðΔxrÞBrow;s
ice (14)
Wealth accumulation
V¼Vt1þYDHS C(15)
The households allocate their wealth in financial assets and receive returns in the form
of interest. The idea behind the portfolio allocation dates back to the work of James Tobin
and his co-workers in the late 1960s in an effort to model basic relationships in a financial
economy with several assets and their allocation among sectors (Brainard/Tobin 1968).
However, our setting slightly differs from the standard practice of modelling financial
assets. Following a more practical approach in Icelands case, the household sector in
Iceland is faced with a decision of allocating its wealth in three financial assets, namely
Icelandic bills ðBh;ice;d
ice Þ, foreign bills ðBh;row;d
ice Þand deposits in Icelandic banks ðDh;ic e;d
ice Þ
operating domestically (equation (16)).
Portfolio allocation of domestic households
Dh;ice;d
ice
Bh;ice;d
ice
Bh;row;d
ice
2
6
6
43
7
7
5¼
λ10
λ20
λ30
2
6
43
7
5Vþ
λ11λ12 λ13
λ21λ22 λ23
λ31λ32 λ33
2
6
43
7
5
rD
ice
rB
ice
rB
row þdxre
2
6
43
7
5(16)
On the other hand, the rest of the world can allocate their wealth in four assets,
that is, Icelandic bills ðBh;ice;d
row Þ,RoWbillsðBh;row;d
row Þ, deposits in the RoW banks
ðBh;row;d
row Þ, and deposits in the foreign branches of Icelandic banks,ðDh;ice;d
row Þas
shownintheportfoliomatrixinequation(17).
17
The portfolio allocation in our
model is based on the fact that Icelandic households were not allowed to hold depos-
its in the foreign branches of Icelandic banks. Clearly this might seem to be at odds
with the principle of free capital movements, but in practice this is precisely what
happened when Icelandic banks were offering foreign depositors considerably higher
rates than to depositors in Iceland.
17. Note that the two assets ðBh;row;d
row Þand ðBh;row;d
row Þin the portfolio allocation are redundant and
do not interact with our model, but are nonetheless included for completeness.
Crises and capital controls in small open economies: a stockflow consistent approach 103
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Portfolio allocation of foreign households
Dh;row;d
row
Dh;ice;d
row
Bh;row;d
row
Bh;ice;d
row
2
6
6
6
6
4
3
7
7
7
7
5
¼
λ50
λ60
λ70
λ80
2
6
6
6
6
4
3
7
7
7
7
5
Vþ
λ51λ52 λ53λ54
λ61λ62 λ63λ64
λ71λ72 λ73λ74
λ81λ82 λ83λ84
2
6
6
6
6
4
3
7
7
7
7
5
rD
row
rD;row
ice
rB
row
rB
icedxre
2
6
6
6
6
6
4
3
7
7
7
7
7
5
(17)
The international traders investing in Icelandic bills also consider their exposure to cur-
rency risks ðdxreÞ, that is, a rise in currency risk would lower the demand for Icelandic
bills. However, the exchange-rate risk in our model does not affect the demand for deposits
in the banks because these deposits are held in the foreign branches of the Icelandic banks.
For example, a UK resident holding a deposit in an Icelandic bank that operates in the UK
will not consider exchange-rate risks on his deposits as the bank account is in GBP.
3.2.3 Banks
The banking sector is the sole provider of credit in the economy. It fulfils the demand for
loans and will only issue bills ðBb;ice;s
ice Þif the demand for loans ðLs
iceÞis greater than the
available credit in the system (equation (19)). As can be seen, the first source of credit
in the economy is provided by the available deposits ðDd
iceÞin the banking sector. This,
however, does not put a restriction on the process of credit creation, that is, if the demand
for loans exceeds the available credit in the banking sector, the banks will issue bills to
finance their operations and fulfil the demand for loans in the economy. As described
earlier, these bills are held by households and the rest of the world, providing credit
and willing to take a counterparty risk on Icelandic banks. Moreover, we do not put a
restriction on the flow of credit between the domestic and foreign branches of the Icelan-
dic banks. This implies that the deposits accumulated in the foreign branches can be used
by the banks domestically without any direct risks involved. The deposits in the foreign
branches of Icelandic banks are also considered as an inflow in the financial account even
if these deposits are not transferred to the domestic banks.
Profit of the banks
Fb¼rL;ice
ice ðLf;ice;dÞþrL;row
ice ðLf;row;dÞrD;row
ice ðDh;ice;d
row ÞrD
iceðDh;ice;d
ice ÞrB
iceðBb;s
ice Þ(18)
Bills issued by the Icelandic banks
ΔBb;ice;s
ice ¼z2ðΔLs
ice ðΔDd
ice þFbÞÞ
z2¼1;if Ls
ice >ðΔDd
ice þFbÞ;otherwise 0 (19)
Icelandic bills issued in the FX market
Beb;s
row ¼Bb;ice;s
ice
Bh;ice;d
ice Bh;ice;d
row ðxrÞBb;ice;d
ice
(20)
It is worth mentioning that traditional models designed for conventional monetary pol-
icy, especially before the crisis, ignored the special role played by the banks in creating
credit from a balance-sheet perspective as discussed above. In general, the banking sector
in these models in itself is not a source of shock and vulnerabilities (Benes et al. 2014).
zfflfflfflfflffl}|fflfflfflfflffl{
bills issued
zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{
bills demanded
104 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
In contrast, based on the balance-sheet approach, the banking sector in our framework can
be a source of vulnerability and shocks by fulfilling all demand for credit due to their opti-
mistic expectations concerning growth prospects. This feature of the financial sector is
widely acknowledged by several authors (see Jakab/Kumhof 2015).
3.2.4 Government
For completeness, we include a simple government sector in the model which finances its
expenditures through taxes. The expenditure of the government is simply determined by
its tax revenue (equations (21)(22)), hence running a balanced budget. This is a strong
assumption but our setting here is based on the Icelandic experience, where the government
sector has a relatively minor or non-existent role in the accumulation of external debt. Ice-
landic firms and banks are generally held responsible for creating economic instability as
they accumulated a large volume of external debt.
Tax revenue
T¼TfþTh(21)
Government budget balance
G¼T(22)
3.2.5 Balance of payments and exchange-rate setting
The current-account balance is presented in equation (23), taking into account net exports
and factor payments in both directions. The determination of trade flows in our model is
based on the standard demand theory. The volumes of exports and imports are presented in
log-linear form (equations (24)(27)) in order to estimate elasticities. Exports and imports in
the model are determined by the level of price competitiveness as well as by the income of
the domestic economy ðyÞand that of the trading partners ðy
rowÞ. Exports are modelled as
partly exogenous, reflecting Icelands natural resource export base and partly dependent on
the real exchange rate, reflecting price-sensitive service exports such as tourism. The level of
the nominal exchange rate affects tradeables through its effects on prices. Equations (28) and
(29) explain the prices of tradeables from the perspective of Iceland in the model. Export
prices ðPxÞare determined by domestic prices ðPyÞ, foreign prices ðPy
rowÞand the exchange
rate ðxrÞ. Import prices which are exogenous are expressed in nominal currency by simply
adjusting the export prices of trading partners for the nominal exchange rate.
Current-account balance
CAB ¼XMþrB
rowðt1ÞðBh;row;d
iceðt1ÞÞrB
iceðt1ÞðBh;ice;d
rowðt1ÞÞxr
rD;row
iceðt1ÞðDh;ice;d
rowðt1ÞÞxr rB
iceðt1ÞðBeb;d
rowðt1ÞÞxr
(23)
Real imports
logðmÞ¼μ0μ1log Pm
Py

þμ2logðyÞ(24)
Real exports: tourism
logðx1Þ¼ϵ0ϵ1log Px
Py
row

þϵ2logðy
rowÞ(25)
Crises and capital controls in small open economies: a stockflow consistent approach 105
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Exports based on natural resources
logðx2Þ¼ϵ0þϵ2logðy
rowÞ(26)
Total exports
x¼α3ðx2Þþ1α3ðx1Þ(27)
Import prices
Pm¼Px
rowðxrÞ(28)
Export prices
logðPxÞ¼νx0þνx1logðPy
rowÞþð1νx1ÞlogðPyÞþνx1logðxrÞ(29)
The nominal exchange rate (equation (30)), based on the portfolio balance approach, is
determined by the ratio of bills issued and demanded in the FX market. During normal
times, the traders consider a fundamental currency risk by taking into account the
exchange-rate misalignment, that is, the deviation of the exchange rate from its long-
run path (equation (32)). This particular measure of risk is based on the model of
Lavoie/Daigle (2011), however those authors consider two types of agents in the FX mar-
ket, speculators and fundamentalists, with fixed weights. Our setting of exchange-rate risks
and expectations differs from those authors. We allow for regime shift where investors
during a sudden stop change their expectations as they anticipate a currency crisis. This
will be discussed in more detail when we introduce a sudden stop in the model.
Exchange-rate setting: ISK per foreign currency
xr ¼Bh;row;d
ice
Bh;ice;d
row
! (30)
Real exchange rate
rxr ¼xr Py
row
Py
 (31)
Exchange-rate expectations
dxre ¼ΩðxrxrÞ
xr (32)
We do not allow deposits in the foreign branches of the banks to affect the exchange-rate
dynamics in the model. Deposits held in foreign branches do not directly affect the currency
unless converted into local currency. This, however, is an internal operation of the banking
system and is not entirely clear from the literature or existing studies. Finally, the real
exchange rate in our model reflects price competitiveness vis-à-vis foreign countries mea-
sured in common currency.
3.3 Simulations
To understand the dynamics of the real and financial sector in general and the role of capi-
tal flows in particular, we numerically solve the model to achieve a baseline scenario. In
obtaining a numerical solution, we use a combination of estimation and calibration with
106 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
the aim of replicating certain key features of the Icelandic economy.
18
We then introduce
two shocks in different time periods to see the response of the economy.
The shocks generated in the model are explained as follows.
3.3.1 Shock 1: increase in real investment
First, we introduce a real investment (gross fixed capital formation) shock where we
increase the real investment by increasing the ηfrom 1 to 1.10 after ten iterations.
This increases the share of real investment in total output, making it a major driver of
output. The increase in investment, however, requires financing through credit, therefore
the economic growth in this sense is finance-led. This scenario is based on the Icelandic
experience of investment boom in the years leading to the crisis. In particular, investment
to GDP increased from 19 per cent in 2002 to 35 per cent in 2006.
3.3.2 Shock 2: increase in interest-rate differentials
In response to the investment boom generated by rapid credit creation, the Icelandic
central bank raised interest rates, which then created profit opportunities for the carry
trade. The carry trade consisted of foreign currency speculators taking positions in the
Icelandic króna, as well as local businesses and also households borrowing in foreign cur-
rency and hence taking an unhedged positioninthecurrencymarket.Usingthisasa
motivation, we introduce an interest-rate differentials shock after 30 iterations by
increasing the Icelandic interest rates while keeping the effect of the investment shock
(shock 1) in the system.
19
We adjust all the interest rates so that the interest rates in
Iceland are higher than the interest rates abroad. In particular, we increase the return
on Icelandic bills from 2 per cent to 3.5 per cent and also increase the interest rate
on loans linked to domestic currency from 4 per cent to 4.5 per cent.
3.4 Discussion
Figure 3 shows the dynamics of the current-account balance and the financial-account bal-
ance for three different scenarios: (i) the baseline scenario; (ii) the scenario after invest-
ment shock; and (iii) the scenario in which investment shock is followed by an interest
differentials shock in the system.
The small open economy in our model is a net borrower and is running a small current-
account deficit and a financial-account surplus in the baseline scenario. In effect the econ-
omy is operating very close to a current-account balance. The economy interacts with the
rest of the world through trade and financial assets. The financial sector issues bills and holds
deposits internationally to fulfil the demand for credit in the economy. An increase in
investment (gross fixed capital formation) will induce credit inflow into the economy
while pushing the current account into further deficit. This scenario reflects a simple sty-
lised fact: if the demand for loans by firms is not fulfilled by domestic credit, banks borrow
18. We include the real sector of the rest of the world in order to determine the dynamics of
wealth and prices. However, we do not explicitly model the financing decisions of the firms and
the banking sector within the rest of the world and treat these variables as exogenous.
19. The increases in interest rates shown in bold type can be seen in Table A2 in Appendix 1.
Crises and capital controls in small open economies: a stockflow consistent approach 107
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
internationally to fulfil the demand for loans in the economy. Increased international bor-
rowing generates a financial-account surplus and a current-account deficit.
If the investment shock is followed by a rise in interest-rate differentials, the current-
account deficit deteriorates further. A rise in interest rate differentials creates an oppor-
tunity for carry trade. The international traders, despite considering exchange-rate
misalignment (that is, the deviation of the exchange rate from its long-run path),
allocate their wealth into Icelandic assets, triggering inflow of credit into the economy.
The introduction of shock 2 in the model allows a speculative attack on the currency.
The deteriorating balance-of-payments position, apart from a speculative attack, is due
to an increase in the cost of borrowing in the economy, which in turn results in even
higher demand for credit. The burden of a higher interest rate in our model is directly
borne by the firms.
We now turn to the real economic indicators in the model. Overall, capital inflows
have a strong positive effect on output, as the fall in trade balance (see Figure 5) is offset
by a rise in domestic demand. Thus, capital inflows are expansionary in nature. In parti-
cular, Figure 9 shows that an increase in real investment boosts real output, as expected.
A rise in investment followed by a rise in the interest-rate differential further increases real
output, as shown in Figure 4. The transmission channel can be explained as follows: a rise
in the interest-rate differential in Iceland (with good sovereign rating) makes Icelandic
assets look more attractive than foreign assets. Thus, households allocate their wealth in
assets with higher returns which in turn increases their wealth as shown in Figure 6.
Thus, the interest-rate differential increases real output through a wealth effect, that is,
an increase in wealth due to higher returns feeds back into output via the consumption
channel in the model, as shown in Figure 8.
Focusing on the dynamics of the exchange rate, Figure 7 shows that a rise in interest-
rate differentials results in the deviation (appreciation in this case) of the exchange rate
from its baseline, which has negative effects on the trade balance. An investment shock
alone (with no direct speculative attack on the currency) has small effects on the
−0.50
−0.25
0.00
0.25
0.50
0.75
0 20 40 60 80 100
Current account/GDP (shock 1 + shock 2)
Current account/GDP (shock 1)
Current account/GDP (baseline)
Financial account/GDP (shock 1 + shock 2)
Financial account/GDP (shock 1)
Financial account/GDP (baseline)
Figure 3 Balance of payments
108 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
64
65
66
67
68
69
0 20 40 60 80 100
Real output (shock 1 + shock 2)
Real output (shock 1)
Real output (baseline)
Figure 4 Real GDP
−0.035
−0.030
−0.025
−0.020
−0.015
−0.010
−0.005
0 20 40 60 80 100
Net exports/GDP (shock 1 + shock 2)
Net exports/GDP (shock 1)
Net exports/GDP (baseline)
Figure 5 Net exports
Crises and capital controls in small open economies: a stockflow consistent approach 109
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
0.500
0.505
0.510
0.515
0.520
0.525
0 20 40 60 80 100
Wealth/GDP (shock 1 + shock 2)
Wealth/GDP (shock 1)
Wealth/GDP (baseline)
Figure 6 Wealth
0.97
0.98
0.99
1.00
1.01
0 20 40 60 80 100
Exchange rate (shock 1 + shock 2)
Exchange rate (shock 1)
Exchange rate (baseline)
Figure 7 Exchange rate
110 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
exchange rate. The asset traders do not change their wealth allocation if the investment
returns on Icelandic and foreign assets are uniform. The mild depreciation in this case is
the result of an increased wealth effect in the model. An increase in Icelandic wealth as
compared to its trading partners will increase the holding of foreign assets by the Icelan-
dic households as they become richer.
0.5600
0.5625
0.5650
0.5675
0.5700
0 20 40 60 80 100
Consumption/GDP (shock 1 + shock 2)
Consumption/GDP (shock 1)
Consumption/GDP (baseline)
Figure 8 Consumption
0.200
0.205
0.210
0.215
0.220
0.225
0.230
0 20 40 60 80 100
Investment/GDP (shock 1 + shock 2)
Investment/GDP (shock 1)
Investment/GDP (baseline)
Figure 9 Investment
Crises and capital controls in small open economies: a stockflow consistent approach 111
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Exchange-rate dynamics in the model also affect the domestic price of goods, that is, a
currency appreciation (depreciation) results in lower (higher) import prices, which
decreases (increases) the domestic price of goods.
Figure 10 and 11 show two important financial indicators in the economy. The effects
of both shocks in the model are clearly reflected in the rising external debt to GDP and
0
5
10
15
0 20 40 60 80 100
Debt/GDP (shock 1 + shock 2)
Debt/GDP (shock 1)
Debt/GDP (baseline)
Figure 10 Debt to GDP
0
5
10
15
0 20 40 60 80 100
M4/GDP (shock 1 + shock 2)
M4/GDP (shock 1)
M4/GDP (baseline)
Figure 11 M4 to GDP
112 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
increased M4 (money supply) to GDP. The dynamics of these stockflow norms follow-
ing an increasing trend are consistent with the actual data of Iceland, where the main
financial indicators relative to GDP follow an upward trend before the crisis. The
increased trend in these financial indicators implies that the growth of the financial sector
is faster than the real growth of the economy.
Figure 12 shows the demand for ISK and foreign-denominated loans by firms. An
increase in the interest-rate differentials will trigger borrowing in foreign-denominated
loans. It can therefore be argued that when the channel of borrowing in foreign currency
is open, as was the case in Iceland, the increase in interest rate is not very effective in con-
trolling domestic demand. Apart from lower interest rates on FX loans, the rise in demand
for FX borrowing could also be the result of currency appreciation as domestic sectors,
such as firms and households, are encouraged to speculate and benefit from exchange-
rate movements.
In the section above, we demonstrated how international financial flows can fuel a
credit boom, which in turn can create an economic boom. This, however, comes at the
cost of persistent current-account deficits and growing external debt as discussed above.
Such scenarios prevailed in many small open economies in the years preceding the crisis.
The situation, however, changed dramatically when a wave of international bankruptcies
affected many small open economies. The sudden stop of capital inflows revealed the fra-
gility and inability of small open economies to deal with the crisis.
3.4.1 Crises and sudden stops
In order to capture the destabilising role of international credit in small economies, we
create a scenario where shock 1 and shock 2 are followed by a sudden stop. We allow
the external debt to GDP to reach six times the size of the economy. Such a high level
of external debt relative to the economy raises concerns and the international creditors
stop lending to the economy. We impose the following restriction on Equation (20):
Change in Icelandic bills issued in the international market and held by the foreign bank
ΔBeb;s
row ¼0;if ðBs
ice=Y
6Þ(20a)
0.35
0.45
0.55
0.65
0 20406080100
Domestic currency loans/total loans (shock 2)
Domestic currency loans/total loans (baseline)
FX loans/total loans (shock 2)
FX loans/total loans (baseline)
Figure 12 Loans breakdown
zfflfflffl}|fflfflffl{
debt to GDP
Crises and capital controls in small open economies: a stockflow consistent approach 113
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
The foreign banks stop holding Icelandic bills and the flow of credit comes to a halt.
20
The firms are unable to finance their investments, resulting in the collapse of real
investments.
After sudden stops, the credit rating of the debt issuers fall, and the international tra-
ders readjust their strategy in the financial markets. The interest-rate differentials, which
in normal times with good credit ratings are seen as an opportunity to make profit, are
seen during the crises as a measure of risk premia. During such circumstances, an increase
in the interest-rate differentials induces capital reversals. Equation (32) can be augmented
as follows:
Changes in expectations during the crisis
dxre ¼ΩðxrxrÞ
xr þðrB
ice rB
rowÞ(32a)
Hence, any further increase in interest-rate differentials would further increase the
expectations of a currency crisis. The anticipation of a currency crisis is exactly what causes
a currency crisis, when international traders start selling their Icelandic assets.
During a sudden stop, the sectors with higher debt are the first ones to go bankrupt as
they fail to pay their liabilities. Bankruptcy from one sector quickly spreads to the other
sectors of the economy. In our model, firms cannot remain solvent even if the level of invest-
ment is significantly reduced and no new borrowing takes place. The reason is that firms
have to pay a higher interest rate on the large pile of existing debt trapped in the economy.
This triggers the balance-of-payments crisis as shown in Figure 13, leaving no option other
than to impose capital controls. The crisis makes the financial sector unable to repay its
debt along with the interest payments on the current account as it goes bankrupt.
−0.3
−0.2
−0.1
0.0
0.1
0.2
0.3
0.4
020 40 60 80 100
Current account/GDP (2 shocks followed by a sudden stop)
Current account/GDP (baseline)
Financial account/GDP (2 shocks followed by a sudden stop)
Financial account/GDP (baseline)
Figure 13 Balance of payments
20. It should be noted here that the flow of credit (inflows) goes to zero which means the stock of
debt will remain constant. The stock of debt will only reduce when an outflow of credit takes place.
114 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
The anticipated currency crisis in the model results in the foreign households selling their
assets in Iceland, resulting in capital reversal to some extent. This market run on the banking
sector generates currency crisis as shown inFigure 15. It is important to highlight that capital
reversal takes the form of foreign households selling their assets in Iceland to a certain limit
after they realise that the foreign financial institutions have stopped lending to the economy.
However, the debt held by foreign financial institutions remains trapped because no one is
able or willing to buy such a large pile of a countrysdebtthatisgoingthroughacrisis.
−6
−4
−2
0
2
4
020 40 60 80 100
Output growth (2 shocks followed by a sudden stop)
Figure 14 Real GDP
0.96
1.02
1.08
1.14 Exchange rate (2 shocks followed by a sudden stop)
Exchange rate (baseline)
020 40 60 80 100
Figure 15 Exchange rate
Crises and capital controls in small open economies: a stockflow consistent approach 115
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
The crisis due to a sudden stopalso results in the collapse of the real economyas shown by
the contraction of output in Figure 14. The level of economic activities greatly reduces,
resulting in a phase of economic recession which in our model is explained by a compression
of domestic demand. Apart from domestic demand compression, as shown in Figures 18
and 19, currency depreciation after the crisis in our model results in a significant reduction
in real imports, improving the trade balance (see Figure 16). This transforms the growth
engine of the economy from investment-led to export-led. In addition, there is a strong
−0.04
−0.02
0.00
0.02
0.04 Net exports/GDP (2 shocks followed by a sudden stop)
Net exports (baseline)
020 40 60 80 100
Figure 16 Net exports
0.49
0.51
0.53
0.55
0.57
0.59 Wealth/GDP (2 shocks followed by a sudden stop)
Wealth/GDP (baseline)
020 40 60 80 100
Figure 17 Wealth
116 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
0.545
0.555
0.565
0.575
0.585
0 20 40 60 80 100
Consumption/GDP (2 shocks followed by a sudden stop)
Consumption/GDP (baseline)
Figure 18 Consumption
0.15
0.17
0.19
0.21
0.23
0.25
0 20 40 60 80 100
Investment/GDP (2 shocks followed by a sudden stop)
Investment/GDP (baseline)
Figure 19 Investment
Crises and capital controls in small open economies: a stockflow consistent approach 117
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
exchange-rate pass-through to prices as shown in Figure 21, which is in line with the experi-
ence of Iceland after the crisis.
21
Our model assumes asymmetric effects regarding exchange-rate dynamics, that is, the
depreciating effect of capital outflow on the currency is much stronger than the appreciat-
ing effect associated with the capital inflows in the years before the crisis. This, as
explained earlier, is due to the shift in the strategy of international investors. This result
is consistent with the Icelandic crisis as well as many other currency crises of the past.
3.4.2 Capital controls and leakages
As shown above, once the crisis hits, the current-account deficit cannot be maintained as
the economy is not able to finance the deficit due to a sudden stop. Consequently, the
economy cannot pay any interest on its debt and has to implement capital controls in
order to prevent any further outflows.
We now add the following conditions to equations (16) and (17):
ΔBh;row;d
ice ¼0
;if ΔBeb;s
row ¼0
(16a)
ΔBh;ice;d
row ¼0
;if ΔBeb;s
row ¼0
(17a)
Figure 22 shows a plausible scenario after the implementation of strong capital controls
in our model. The country is not able to pay any interest on the external debt trapped in
1
3
5
7
9
0 20 40 60 80 100
Debt/GDP (2 shocks followed by a sudden stop)
Debt/GDP (baseline)
Figure 20 Debt
zfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflffl{
sudden stop
zfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflffl{
captial controls
zfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflffl{
captial controls
zfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflffl{
sudden stop
21. The stock effects of a sudden stop can be seen in Figures 20 and 17, showing the dynamics of
debt and household wealth in the economy respectively.
118 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
1.065
1.075
1.085
1.095
1.105
1.115
0 20 40 60 80 100
Sales price (2 shocks followed by a sudden stop)
Sales price (baseline)
Figure 21 Domestic price of goods
Current account/GDP (2 shocks followed by a sudden stop and capital controls)
Current account/GDP (baseline)
Financial account/GDP (2 shocks followed by a sudden stop and capital controls)
Financial account/GDP (baseline)
−0.3
−0.2
−0.1
0.0
0.1
0.2
0.3
0.4
020 40 60 80 100
Figure 22 Balance of payments
Crises and capital controls in small open economies: a stockflow consistent approach 119
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
the economy as a wave of bankruptcies propagates through the balance sheets of all the
sectors in the economy. This situation forces the economy to implement capital controls
immediately in order to restrict outflows and prevent a currency crisis.
The situationin Figure 22 is based on a proactive response by the authorities, where capi-
tal controls with zero leakages are implemented before the outflow has occurred. In reality,
international traders are quicker to sell their assets before the authorities can respond. Thus,
implementation of capital controls can never fully prevent outflows. Moreover, tighter capi-
tal controls with zero leakages are practically impossible, and international traders will take
their capital out of the economy whenever there is an opportunity. Capital controls, how-
ever, can slow down the outflow of international capital.
We introduce capital leakages into our model by allowing a small fraction of for-
eign capital to escape in every period. Figure 23 shows the dynamics of exchange rates
after capital leakages in every period. If the capital continues to leak, the exchange
rate will continuously depreciate, as shown in Figure 23. The gradual depreciation
duetoleakagesisstilllessharmfulthanafull-blown currency crisis taking place in
a very short period of time. In our model, small leakages have some effect on the cur-
rency but in reality small leakages may or may not put any pressure on the currency.
Capital controls give a country enough time to restructure the financial structure and
stabilise the economy. During a strong capital control regime or even with low leakages,
the country can gradually lower the interest rates which can restore domestic demand
by channelling savings into investment, having little or no impact on the currency, as
was the case in Iceland. Thus, the economy can retain export-led growth and ease the
burden of interest repayments on the current account. The trade surplus in the econ-
omy can be used to repay its debt. The effects of lowering interest rates, as opposed to
the IMFs insistence on keeping them high under a capital control regime in the case of
Iceland, are discussed in Gudmundsson/Zoega (2016).
22
0.96
1.02
1.08
1.14
0 20 40 60 80 100
Exchange rate (capital leakages)
Exchange rate (strong capital controls)
Exchange rate (baseline)
Figure 23 Exchange rate
22. The authors provide an extensive discussion on the effects of lowering interest rates under a
capital control regime in Iceland.
120 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
We now turn to the sensitivity analysis of our model by changing the values of key
parameters. We focus on the parameter values determining the balance of payments
and exchange-rate dynamics in the model. After changing the parameter values, we simu-
late the model for the two shocks an investment shock followed by an interest rates dif-
ferential shock defined in Section 3.3.
As expected, changes in key parameters lead to different steady states. However, the
effects of the shocks and the implications of our original model do not change in any funda-
mental way, that is, the simulations of the system still reveal the same dynamics.
23
The only
major impact is observed in the case of exchange rates which achieve a very different steady
state in response to a change in the portfolio allocation, but the effects of the shocks are
still in line with our main analysis. Overall, we can argue that our model is not sensitive to
changes in the values used in the baseline.
4 CONCLUSION
This paper proposed a framework for a small open economy with sovereign currency while
focusing on the role of international financial flows. We demonstrated that borrowing to
finance real investment under lower interest rates has a mild impact on the current-
account balance and a stronger impact on the real output. In contrast, international bor-
rowings in a small open economy with sovereign currency often require offering better
incentives to international investors than its competitors due to exchange-rate risks.
This, in Iceland, took the form of higher interest returns on Icelandic financial assets rela-
tive to foreign assets. In this regard, our analyses suggested that positive interest-rate dif-
ferentials attract a large amount of short-term inflows by creating an opportunity for
profits. These persistent inflows are found to overvalue the exchange rate, deteriorate
the trade balance and generate an economic boom with a fast growing debt.
The accumulation of external debt may or may not reach unsustainable levels, but
nonetheless will at some point create concerns in the markets regarding the potential
of debt repayments. These market fears take the form of a sudden stop as international
short-term capital, due to its fleeting nature, flows out of the economy in a very short
period of time. These sudden outflows convert liquidity crises into solvency crises and
sustainable debts into unsustainable debts, creating serious financial and real economic
crises.
A sudden outflow severely impacts the whole economy. It generates a balance-of-payments
crisis and also compresses domestic demand, resulting in a severe recession. A small open
economy reliant on international credit is not able to finance its economic activities after
the channel of international capital is closed. An appropriate immediate response in the
short run is to impose strong capital controls in order to stabilise the currency and gain enough
time to restructure the economy. To facilitate the balance-of-payments adjustment and a
quick economic recovery, capital controls require the backing of careful monetary policy deci-
sions. The cost of borrowing at the time of the crisis skyrockets, and a financial crisis usually
leaves the country to operate in an environment of high interest rates due to fears of inflation
and currency crisis. This further triggers the paradox of thrifteffect with a rise in savings and
a considerable decline in real investments due to economic uncertainties and the heavy cost
of borrowing.
Monetary authorities can gradually reduce interest rates to recover domestic demand in
order to facilitate economic growth. In a capital control regime, interest rates can have
23. These results are not reported, to conserve space, however they can be provided upon request.
Crises and capital controls in small open economies: a stockflow consistent approach 121
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
weak or no significant effects on the currency as the channel of trading financial assets is
restricted. Thus, lowering interest rates will not affect export-led growth in the capital con-
trol regime but will ease the burden of interest payments to international creditors, further
improving the balance of payments. This argument in Icelands particular case is sup-
ported by Gudmundsson/Zoega (2016). The removal of capital controls, however,
remains a challenge for small open economies going forward. The effects of relaxation
in capital controls, and the future strategy regarding capital inflows in Iceland, are dis-
cussed in Zoega (2016).
REFERENCES
Accominotti, O., Eichengreen, B. (2016): The mother of all sudden stops: capital flows and reversals
in Europe, 191932, in: The Economic History Review, 69(2), 469492.
Baldursson, F.M., Portes, R. (2013): Gambling for resurrection in Iceland: the rise and fall of the
banks, CEPR Working Paper No 9664.
Benediktsdottir, S., Danielsson, J., Zoega, G. (2011): Lessons from a collapse of a financial system,
in: Economic Policy, 26(66), 183235.
Benes, J., Kumhof, M., Laxton, D. (2014): Financial crises in DSGE models: a prototype model,
IMF Working Paper, No 14/57.
Bordo, M.D., Landon-Lane, J. (2013): Does expansionary monetary policy cause asset price booms:
some historical and empirical evidence, NBER Working Paper No 19585.
Brainard, W.C., Tobin, J. (1968): Pitfalls in financial model building, in: The American Economic
Review, 58(2), 99122.
Burgess, S., Burrows, O., Godin, A., Kinsella, S., Millard, S. (2016): A dynamic model of financial
balances for the United Kingdom, Bank of England Staff Working Paper No 614.
Byrialsen, M.R., Raza, H. (2018): Macroeconomic effects of unemployment benefits in small open
economies: a stockflow consistent approach, in: European Journal of Economics and Economic
Policies: Intervention, 15(3), 335363.
Caiani, A., Catullo, E., Gallegati, M. (2018): The effects of fiscal targets in a monetary union: a
multi-country agent-based stock flow consistent model, in: Industrial and Corporate Change,
27(6), 11231154.
Calvo, G.A., Reinhart, C.M. (1999): When capital inflows come to a sudden stop: consequences
and policy options, in: Kenen, P.B., Swoboda, A.K. (eds), Reforming the International Monetary
and Financial System, Washington, DC: IMF, 175201.
Caverzasi, E., Godin, A. (2015): Post-Keynesian stockflow-consistent modelling: a survey, in:
Cambridge Journal of Economics, 39(1), 157187.
Dornbusch, R. (1975): Exchange rates and fiscal policy in a popular model of international trade, in:
The American Economic Review, 65(5), 859871.
Duwicquet, V., Mazier, J. (2010): Financial integration and macroeconomic adjustments in a mone-
tary union, in: Journal of Post Keynesian Economics, 33(2), 333370.
Eichengreen, B.J. (2004): Capital Flows and Crises, Cambridge, MA: MIT Press.
Engel, C., Matsumoto, A. (2006): Portfolio choice in a monetary open-economy DSGE model,
Technical report, National Bureau of Economic Research.
Farhi, E., Werning, I. (2014): Dilemma not trilemma? Capital controls and exchange rates with
volatile capital flows, in: IMF Economic Review, 62(4), 569605.
Ghosh, A.R., Qureshi, M.S. (2016): Whats in a name? That which we call capital controls, IMF
Working Paper No WP/16/25.
Ghosh, A.R., Ostry, J.D., Qureshi, M.S. (2017): Taming the Tide of Capital Flows: A Policy Guide,
Cambridge, MA: MIT Press.
Godley, W., Lavoie, M. (2007): Monetary Economics: An Integrated Approach to Credit, Money,
Income, Production and Wealth, London: Palgrave Macmillan.
Greenwood-Nimmo, M. (2014): Inflation targeting monetary and fiscal policies in a two-country
stockflow-consistent model, in: Cambridge Journal of Economics, 38(4), 839867.
122 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Gudmundsson, B.R. (2016): Financialisation and financial crisis in Iceland, in: European Journal of
Economics and Economic Policies, 13(3), 292322.
Gudmundsson, G.S., Zoega, G. (2016): A double-edged sword: high interest rates in capital control
regimes, in: The Open-Access, Open-Assessment E-Journal, 10(201617), 138.
Harvey, J.T. (2009): Currencies, Capital Flows and Crises: A Post Keynesian Analysis of Exchange Rate
Determination, London: Routledge.
Jakab, Z., Kumhof, M. (2015): Banks are not intermediaries of loanable funds and why this mat-
ters, Bank of England Working Paper, No 529.
Kinsella, S., Khalil, S. (2012): Debt-deflation in a stockflow consistent macromodel, in: Papadimi-
triou, D.B., Zezza, G. (eds), Contributions in StockFlow Consistent Modeling: Essays in Honor of
Wynne Godley, London: Palgrave Macmillan, 235265.
Korinek, A., Sandri, D. (2016): Capital controls or macroprudential regulation?, in: Journal of Inter-
national Economics, 99, S27S42.
Krugman, P. (2014): Currency regimes, capital flows, and crises, in: IMF Economic Review, 62(4), 470493.
Laeven, M.L., Valencia, F. (2010): Resolution of banking crises: the good, the bad, and the ugly,
IMF Working Paper No 10/146.
Lane, P.R., McQuade, P. (2014): Domestic credit growth and international capital flows, in: The
Scandinavian Journal of Economics, 116(1), 218252.
Lavoie, M., Daigle, G. (2011): A behavioural finance model of exchange rate expectations within a
stockflow consistent framework, in: Metroeconomica, 62(3), 434458.
Magud, N.E., Reinhart, C.M., Rogoff, K.S. (2011): Capital controls: myth and reality a portfolio
balance approach, Technical Report, National Bureau of Economic Research.
Matthiasson, T. (2008): Spinning out of control: Iceland in crisis, in: Nordic Journal of Political
Economy, 34(3), 119.
Mazier, J., Tiou-Tagba Aliti, G. (2012): World imbalances and macroeconomic adjustments:
a three-country stockflow consistent model with fixed or flexible prices, in: Metroeconomica,
63(2), 358388.
Meijers, H., Muysken, J., Sleijpen, O. (2015): The deposit financing gap: another Dutch disease, in:
European Journal of Economics and Economic Policies, 12(1), 3250.
Nikiforos, M., Zezza, G. (2017): Stockflow consistent macroeconomic models: a survey, in: Jour-
nal of Economic Surveys, 31(5), 12041239.
Obstfeld, M. (2012): Financial flows, financial crises, and global imbalances, in: Journal of Interna-
tional Money and Finance, 31(3), 469480.
Raza, H., Gudmundsson, B., Zoega, G., Kinsella, S. (2016): Two thorns of experience: financialisa-
tion in Iceland and Ireland, in: International Review of Applied Economics, 30(6), 771789.
Raza, H., Zoega, G., Kinsella, S. (2018): Capital inflows, crisis and recovery in small open econo-
mies, in: Finance Research Letters, 27, 273282.
Reinhart, C.M., Reinhart, V.R. (2008): Capital flow bonanzas: an encompassing view of the past
and present, NBER Working Paper No 14321.
Rodrik, D. (2000): How far will international economic integration go?, in: The Journal of Economic
Perspectives, 14(1), 177186.
Schularick, M., Taylor, A.M. (2009): Credit booms gone bust: monetary policy, leverage cycles and
financial crises, 18702008, NBER Working Paper No 15512.
Zeev, N.B. (2017): Capital controls as shock absorbers, in: Journal of International Economics, 109,
4367.
Zoega, G. (2016): Responding to capital flows in a very small economy, in: Atlantic Economic Jour-
nal, 44(2), 159170.
Crises and capital controls in small open economies: a stockflow consistent approach 123
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Table A1 Transaction flow matrix (TFM)
Small economy RoW
Flows Households Firms Govt Banks Foreign branch Households Firms RoW banks
Current Capital
Consumption CþC––
Investment þII––
Govt exp. þGG––
Exports þX–– Mrow
Imports M–– þXrow
(GDP) ½Y––
Wages þWB WB ––
Tax ThTfþT––
Interest on deposits in
Icelands domestic
banks
þrD
iceðDh;ice;d
ice Þ––
rD
iceðDh;ic e;d
ice Þ––
Interest on deposits in
Icelands foreign banks
––
rD;row
ice ðDh;ice;d
row ÞþrD;row
ice ðDh;ice;d
row Þ––
Interest on deposits in
RoW banks
––––
þrD
rowðDh;row;d
row ÞrD
rowðDh;row;d
row Þ
Interest on FX denomi-
nated loans in Iceland
rL;rowðLf;row;dÞ––
þrL;rowðLrow;dÞ––
Interest on lSK denomi-
nated loans in Iceland
rL;ice
ice ðLf;ice;dÞ––
þrL;ice
ice ðLice;dÞ––
Interest on Icelandic
banksbills þrB
iceðBice;d
ice Þ––
rB
iceðBb;s
ice ÞþrB
iceðBh;ice;d
row ÞþrB
iceðBeb;b;d
row Þ
Interest on RoW bills þrB
rowðBh;row;d
ice Þ––
þrB
rowðBh;row;d
row ÞrB
rowðBf;s
rowÞ
Profits (firms) FfþFf––
APPENDIX 1
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Change in Icelandic
banksbills
ΔBice;d
ice ––
þΔBb;s
ice ΔBh;ice;d
row ΔBb;s
ice
Change in RoW bills ΔBh;row;d
ice ––
ΔBh;row;d
row þΔBf;s
row
Change in IcelandsFX
denominated loans
––
þΔLf;row;dΔLrow;s––
Change in Icelands ISK
denominated loans
––
þΔLf;ice;dΔLice;s––
Change in deposits in
Icelandic domestic
banks
ΔDh;ice;d
ice ––
þΔDh;ice;d
ice ––
Change in deposits in
Icelandic foreign banks
––
þΔDh;ice;d
row ΔDh;ice;d
row ––
Change in deposits in
foreign banks
––––
ΔDh;row;d
row þΔDh;row;d
row
0000 0
Note: Rest of the world is largely exogenous. In our transaction flow matrix, we only report those flows through which the domestic economy is connected to the rest of the
world. The columns for the rest of the world sector may not sum to zero.
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
A1.1 Model equations
A1.1.1 Firms
Gross domestic product (GDP)
Y¼CþIþGþXM
Sales in Iceland
S¼CþIþGþX
Value of real output
y¼cþiþgþxm
GDP deflator
Py¼Y=y
Employment
N¼y=A
Wage bill
WB ¼WðNÞ
Unit cost
UC ¼ðWB þMÞ=s
Price of domestic goods
Ps¼ð1þϕÞðUCÞ
Domestic sales price
Pds ¼SX
sx
Real sales
s¼cþgþiþx
Real investment
i¼γðkTkt1Þþda
Targeted stock of capital
kT¼ηðyt1Þ
Depreciation of capital
da ¼δðkt1Þ
Change in stock of capital
Δk¼ida
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
126 European Journal of Economics and Economic Policies: Intervention, Vol. 16 No. 1
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Nominal value of investment
I¼iðPdsÞ
Nominal value of sales
S¼sðPsÞ
Profit of the firms
Ff¼YWB Tf
Tf¼FfðθÞ
Demand for loans
Łf;d¼Lf;d
t1þIFfþrL;ice
t1ðLf;ice;d
t1ÞþrL;row
t1ðLf;row;d
t1Þ
Demand for ISK and foreign denominated loans
Lf;ice;d¼Lf;dðψ1ω2rL;ice
ice þψ3rL;row
ice Þ
Lf;row;d¼Lf;dðψ4þψ5rL;ice
ice ψ6rL;row
ice Þ
A1.1.2 Households
Households disposable income
YD ¼WB þrB
iceðt1ÞðBice;d
iceðt1ÞÞþrB
rowðt1ÞðBrow;d
iceðt1ÞÞþrD
iceðt1ÞðDh;ice;d
iceðt1ÞÞTh
Th¼YDðθÞ
HaigSimons disposable income
YDHS ¼YD þðΔxrÞBrow;s
ice
Wealth accumulation
V¼Vt1þYDHS C
Real HaigSimons disposable income
ydhs ¼YD
Pds vt1ðΔPdsÞ
Pds
Real wealth
v¼V=Pds
Consumption function
c¼α1ðydhsÞþα2ðvt1Þ
Real consumption
C¼cðPdsÞ
Crises and capital controls in small open economies: a stockflow consistent approach 127
© 2019 The Author Journal compilation © 2019 Edward Elgar Publishing Ltd
Downloaded from PubFactory at 06/30/2023 07:12:53PM
via free access
Portfolio allocation:
Portfolio allocation of domestic households
Dh;ice;d
ice
Bh;ice;d
ice
Bh;row;d
ice
2
6
6
43
7
7
5¼
λ10
λ20
λ30
2
6
43
7
5Vþ
λ11 λ12 λ13
λ21 λ22 λ23
λ31 λ32 λ33
2
6
43
7
5
rD
ice
rB
ice
rB
row þdxre
2
6
43
7
5
Portfolio allocation of foreign households
Dh;row;d
row
Dh;ice;d
row
Bh;row;d
row
Bh;ice;d
row
2
6
6
6
6
4
3
7
7
7
7
5
¼
λ50
λ60
λ70
λ80
2
6
6
6
6
4
3
7
7
7
7
5
Vþ
λ51 λ52 λ53 λ54
λ61 λ62 λ63 λ64
λ71 λ72 λ73 λ74
λ81 λ82 λ83 λ84
2
6
6
6
6
4
3
7
7
7
7
5
rD
row
rD;row
ice
rB
row
rB
icedxre
2
6
6
6
6
6
4
3
7
7
7
7
7
5
A1.1.3 Government
Tax revenue
T¼ThþTf
Government budget balance
G¼T
A1.1.4 Banking sector
Icelandic banks:
Profit of the banks
Fb¼rL;ice
ice ðLf;ice;dÞþrL;row
ice ðLf;row;dÞ
rD;row
ice ðDh;ice;d
row ÞrD
iceðDh;ice;d
ice ÞrB