notes on global imbalances and the low interest rate puzzle
TRANSCRIPT
![Page 1: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/1.jpg)
Notes on Global Imbalances and the LowInterest Rate Puzzle
Vu T. Chau
May 7, 2019
![Page 2: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/2.jpg)
Global Imbalances questions
Related questions that need a joint answer:
Why is the real interest rate so low in the US and other
advanced economies?
Why does capital flow from developing economies to
advanced ones (south-to-north)?
What determines the safe/risky portfolio composition of a
country’s balance sheet?
![Page 3: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/3.jpg)
Outline
Stylized facts.
Benchmark: 2-country neoclassical growth model.• Lucas (1990), Gourinchas and Rey (2014)• Metzler diagram
Heterogeneous demographics and financial frictions. Coeurdacieret al. (2015)
Asset supply explanation. Caballero et al. (2008)
Asset demand explanation. Mendoza et al. (2009)
![Page 4: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/4.jpg)
Stylized Facts
![Page 5: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/5.jpg)
Stylized facts
1 Real interest rates of advanced economies have been declining
since mid-1980s.
2 The US has persistently run current account deficit.
• But net foreign asset declines much less due to valuation effect.
3 Composition: US net holder of world risky assets, ROW net
holder of safe assets.
![Page 6: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/6.jpg)
Fact 1: declining real interest rates
Figure: Short Term Real Interest Rate, U.S., U.K., Germany and France
(G4), 1920-2011
Figure from Gourinchas and Rey (2014).
Real interest rate was sharply declining 1930-1945 too.
![Page 7: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/7.jpg)
Fact 2: US persistent CA deficit
Figure: US persistently runs current account deficit. But its NFA declines
by less.
Figure from Gourinchas and Rey (2014).
![Page 8: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/8.jpg)
Figure: US persistently runs current account deficit. But its NFA declines
by less.
Figure from Gourinchas and Rey (2014).
![Page 9: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/9.jpg)
Fact 3: US long position in risky assets
Figure: Advanced economies long in risky assets. Emerging markets short
risky assets.
Figure from Gourinchas and Rey (2014).
![Page 10: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/10.jpg)
Private vs. Public flows
Figure: Flows to rich economies mainly driven by public flows.
Figure from Aguiar and Amador (2009)
![Page 11: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/11.jpg)
Benchmark: NeoclassicalGrowth Model
![Page 12: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/12.jpg)
Closed-Economy Neoclassical Growth Model I
Closed economy:
max{Ct ,Kt }
∞t=0
∞∑t=0
βtNtu(Ct/Nt)
s.t Ct + Kt+1 = F (Kt ,AtNt) + (1 − δ)Kt
F (Kt ,AtNt) = Kαt (AtNt)
1−α
At and Nt grow at rate gAt and gN
t respectively.
![Page 13: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/13.jpg)
Closed-Economy Neoclassical Growth Model II
Rewrite the problem in effective units of labor:
max∞∑t=0
βtNtu(ctAt)
subject to
ct + kt+1(1 + gAt+1)(1 + gN
t+1) = kαt + (1 − δ)kt
with xt ≡ Xt/(AtNt).
![Page 14: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/14.jpg)
Closed-Economy Neoclassical Growth ModelIII
The economy governed by two equations:
ct + kt+1(1 + gAt+1)(1 + gN
t+1) = kαt + (1 − δ)kt (1)
1 = β(1 + rAutt+1 )u ′(ct+1At+1)
u ′(ctAt)(2)
for rAutt ≡ f ′(kt) − δ.
In transition: low kt imply high MPK, thus high autarky rate.
Capitals should flow from capital-abundant country to
capital-scarce ones.
![Page 15: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/15.jpg)
Detour: Lucas (1990)
Back of envelope calculation:
y = Akα → MPK = r = αA1α y1−
1α
Assuming same A and α = 0.4, then:
r India
rUS=
(y India
yUS
)1− 10.4
= (15)1.5 = 58
Free capital mobility should imply massive capital flowing
from the US to India!
But, previously, we showed an economy in effective units of
labor...
![Page 16: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/16.jpg)
Detour: Lucas (1990)
Lucas does adjust for that: (y instead of y)
Human capital difference.
• Let h denotes human capital, and production is
y = Ahkα → y = Akα.• Normalize y = y/h, so
r India/rUS = (y India/yUS)−1.5 = (1/3)−1.5 = 5.• Still large.
External effect of human capital: y = Akαhζ.
• Use differential in schooling, can get r India/rUS = 1.• Work well across countries.• But this means zero international knowledge spillovers.
Political risks and contracting friction.
• Recall the risk of expropriation from the reputation literature.• But returns were not equalized when India was under British
control before 1945.
![Page 17: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/17.jpg)
Autarky steady state
Steady state: pair (c , k) constant that satisfy 1 and 2.
Assume CRRA utility.
β(1 + rAut)A−γt+1
A−γt
= 1 ⇒ rAut =1
β(1 + gA)γ − 1
With β = e−ρ, can approximate
rAut = ρ+ γgA
Economies that grow faster have higher autarky interest rate.
Due to high desire to borrow from the (richer) future.
No role for demographics change here (gN not determinant ofautarky rate).
• Contrast with Coeurdacier et al. (2015) later.
![Page 18: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/18.jpg)
Benchmark: Financial integration
The budget constraint under financial integration
ct+(kt+1+bt+1)(1+gAt+1)(1+gN
t+1) = kαt +(1−δ)kt+(1 + rWt )bt
where bt denotes Home NFA at the beginning of period.
Can show via FOCs that under financial integration, returns
are equalized
rWt = rt = r∗t ∀t
where rt ≡ αkα−1t − δ, r∗t = αk∗α−1
t − δ. Thus:
kt = k∗t ∀t
Exercise: how about consumption growth rate of Home
compared to Foreign?
![Page 19: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/19.jpg)
Assume gA < g∗A, so that rAut < r∗,Aut . Can show through
algebra:
rAut < r < r∗,Aut
gA < gWcons < g∗A
b = −b∗
Slow (fast) technological growth countries have low (high)
autarky interest rate, export (import) capital, and accumulate
foreign assets (liabilities).
![Page 20: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/20.jpg)
Metzler diagram
Figure: Metzler diagram of financial integration
![Page 21: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/21.jpg)
Remarks on Benchmark I
Which countries have high autarky interest rate? Emerging or
developed economies?
In the previous analysis: developing countries grow faster,
thus should have higher autarky interest rate.
But to get capital flow South-North like the data, developing
countries must have lower autarky interest rate (at least from
a theory perspective).
![Page 22: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/22.jpg)
Remarks on Benchmark II
Developing countries can have lower autarky interest ratebecause:
• Large financial friction depresses ability to borrow by young
people. Coeurdacier et al. (2015)• Low capacity to supply asset (instruments for saving) depresses
interest rate. Caballero et al. (2008)• Low financial friction depresses risk-sharing, increasing demand
for precautionary saving, depresses interest rate. Mendoza
et al. (2009)
This class of models all point to financial development as a
key factor, though the channels are somewhat different.
![Page 23: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/23.jpg)
Models Generating Low InterestRate in Developing Countries
![Page 24: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/24.jpg)
Coeurdacier, Guibaud, and Jin (2015)
Original model with capital accumulation. Presented here the
2010D edition.
Overlapping generations: young (y), middle (m), old (o).
Utility of agent born at time t:
Ut = u(cy ,t) + βu(cm,t+1) + β2u(co,t+2)
One-period risk-free bond: cohort t holds ay ,t+1 in period t
and am,t+2 in period t + 1.
BC:
cy ,t + ay ,t+1 = wy ,t
cm,t+1 + am,t+2 = wm,t+1 + Rt+1ay ,t+1
co,t+2 = Rt+2am,t+2
(old people have no income)
![Page 25: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/25.jpg)
No agent can borrow more than fraction θ of NPV future
income
ay ,t+1 ≥ −θwm,t+1
Rt+1
am,t+2 ≥ 0
Production:
Yt = At(etLy ,t + Lm,t)
wy ,t = etAt , wm,t = At
et : modelling device to make young poorer than middle-aged,
thus incentive to borrow when young.
![Page 26: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/26.jpg)
Optimal am,t+1:
u ′(cm,t) = βRt+1u′(co,t+1)
The Euler constraint does not bind for m. (Why?)
Optimal ay ,t+1:
u ′(cy ,t) = βRt+1u′(cm,t+1) + µy ,t
µy ,t
(ay ,t+1 + θ
wm,t+1
Rt+1
)= 0
![Page 27: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/27.jpg)
Middle-aged have to save for retirement. (Asset demand)
Young want to borrow, but are constrained. (Asset supply)
The more severe the financial friction, the lower asset supply
is. This bids up price, and depresses return.
![Page 28: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/28.jpg)
Let 1 + gL,t ≡ Ly,tLm,t
, and 1 + gA,t+1 ≡ At+1
At. With σ = 1:
Rt+1 =1 + β
β
θ
1 − θ(1 + gL,t)(1 + gA,t+1)
Factors that lower supply relative to demand for assets: lower
population growth, lower productivity growth, and more
severe financial friction. (Why?)
![Page 29: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/29.jpg)
Financial integration
Now: two countries U and C . They are financially integrated.
U grows slower than C
(1 + gUL )(1 + gU
A ) < (1 + gCL )(1 + gC
A ) (3)
But C has terrible financial market θC < θU that overall
θC
1 − θC(1 + gC
L )(1 + gCA ) <
θU
1 − θU(1 + gU
L )(1 + gUA ) (4)
(3) is as in benchmark. (4) is the modification to explain the
facts.
![Page 30: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/30.jpg)
World interest rate
Integrated world interest rate:
RWt+1 = (1 − λCt )R
U + λCt RC
World interest rate weighted average of autarky rate.
C has a lower autarky rate.
The world interest rate declines because C becomes more
important in the world financial market:
dλCtdt
> 0
![Page 31: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/31.jpg)
Capital flow
Net foreign asset position of C :
NFACt =
β
1 + β(1 − θC )AC
t LCm,t
(1 −
RC
RWt+1
)
Since RWt+1 > RC ∀t, NFAC
t > 0 ∀t.
Long-run NFA of C grows at rate
NFACt+1
NFACt
t→∞−−−→ (1 + gCA )(1 + gC
L )
![Page 32: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/32.jpg)
Numerical simulation:
Figure: World interest rate over time
Parameters based loosely on Coeurdacier et al. (2015): gUA = 1%,
gCA = 4%, gU
L = 0, gCL = 2%, θU = 0.34, θC = 0.32, β = 0.91
(annually), LCm,0 = 6, LUm,0 = 1, AC0 = 0.1, AU
0 = 3.
![Page 33: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/33.jpg)
Caballero, Farhi, Gourinchas (2008)
Low financial development = inability to securitize available
current and future outputs.
Low asset supply depresses autarky interest rate.
Non-Ricardian equivalence feature is important.
• Lower ability to securitize output depresses supply of asset.• But this increases non-securitizable output, which increases
saving (demand for asset) today if agents are Ricardian.• In equilibrium: level of financial development does not matter
for autarky interest rate.
![Page 34: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/34.jpg)
Model
Continuous-time.
Continuum agents of mass 1.
Birth rate = death rate = θ.
At birth, agents receive endowment (1 − δ)Xt . Save entirelyand consume at time of death.
• No meaningful consumption-saving decision for parsimony.
Vehicle to save: Lucas tree with dividend δXt and value Vt .
δ: degree of financial development.
• δ = 1: all output in the economy can be written as claims and
traded.• δ = 0: no output can be written as claims.• δ important for determining interest rate.
![Page 35: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/35.jpg)
Financial wealth evolve according to
_Wt
Wt︸︷︷︸agg. wealth growth
= rt︸︷︷︸investment return
− θ︸︷︷︸consumed
+ (1 − δ)Xt
Wt︸ ︷︷ ︸noncapitalizable wealth
Return from saving in the Lucas tree:
rt = δXt
Vt︸︷︷︸dividend rate
+_Vt
Vt︸︷︷︸capital gain
![Page 36: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/36.jpg)
In equilibrium: savings must be equaled to the value of the trees:
Wt = Vt
This yields:
Wt =Xt
θ
rAutt =_Xt
Xt+ δθ ≡ g + δθ
Higher g : increases high capital gain, raising interest rate.
Higher δ: more asset supply, raising interest rate.
Higher θ: lower financial wealth, reducing demand for asset,
raising interest rate.
![Page 37: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/37.jpg)
PVt =
∫∞t
e−∫st rτdτXsds
Vt = δPVt
Nt = (1 − δ)PVt
Note that rt endogenous → changing Vt .
Non-Ricardian equivalence is important because δ ↑→ Vt ↑,
but δ ↑→ Nt ↓ one-for-one, which leads to Wt ↑ one for one.
rt does not change.
Customers have full rights over Vt , but not Nt .
![Page 38: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/38.jpg)
Small Open Economy
Define the Current Account and Trade Balance:
CAt = _Wt − _Vt
TBt = Xt − θWt
Vt
Xt= δ
∫∞t
e−r(s−t)Xs
Xtds
t→∞−−−→ δ
r − g
Wt
Xt= W0e
(r−θ)t +
∫ t0(1 − δ)Xse
(r−θ)(t−s)dst→∞−−−→ 1 − δ
g + θ− r
![Page 39: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/39.jpg)
Metzler diagram
![Page 40: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/40.jpg)
Current Account and Trade Balance:
CAt
Xt→ −g
raut − r
(g + θ− r)(r − g)
TBt
Xt→ raut − r
g + θ− r
If raut > r , the SOE runs persistent positive trade balance, but
not enough to service external liabilities.
Thus, running persistent current account deficit.
Got this intuition from benchmark: high autarky interest rate
country imports capital and have net external liabilities.
![Page 41: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/41.jpg)
Mendoza, Quadrini, Rıos-rull (JPE 2009)
Focus on asset demand.
Low financial development = endogenous incomplete market
= inability to self-insure against idiosyncratic risks.
High demand for precautionary saving → low autarky interest
rate.
With risky investment shock, can also explain variations in
portfolio composition between countries.
![Page 42: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/42.jpg)
![Page 43: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/43.jpg)
Bibliography I
Mark Aguiar and Manuel Amador. Growth in the shadow of
expropriation. Working Paper 15194, National Bureau of
Economic Research, July 2009. URL
http://www.nber.org/papers/w15194.
Ricardo J. Caballero, Emmanuel Farhi, and Pierre-Olivier
Gourinchas. An equilibrium model of ”global imbalances” and
low interest rates. American Economic Review, 98(1):358–93,
March 2008. doi: 10.1257/aer.98.1.358. URL http:
//www.aeaweb.org/articles?id=10.1257/aer.98.1.358.
Nicolas Coeurdacier, Stephane Guibaud, and Keyu Jin. Credit
constraints and growth in a global economy. American Economic
Review, 105(9):2838–81, September 2015. doi:
10.1257/aer.20130549. URL http:
//www.aeaweb.org/articles?id=10.1257/aer.20130549.
![Page 44: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/44.jpg)
Bibliography II
Pierre-Olivier Gourinchas and Helene Rey. Chapter 10 - external
adjustment, global imbalances, valuation effects. In Gita
Gopinath, Elhanan Helpman, and Kenneth Rogoff, editors,
Handbook of International Economics, volume 4 of Handbook of
International Economics, pages 585 – 645. Elsevier, 2014. doi:
https://doi.org/10.1016/B978-0-444-54314-1.00010-0. URL
http://www.sciencedirect.com/science/article/pii/
B9780444543141000100.
Robert E. Lucas. Why doesn’t capital flow from rich to poor
countries? The American Economic Review, 80(2):92–96, 1990.
ISSN 00028282. URL
http://www.jstor.org/stable/2006549.
![Page 45: Notes on Global Imbalances and the Low Interest Rate Puzzle](https://reader031.vdocuments.mx/reader031/viewer/2022011813/61d49d32f813ff1143300b81/html5/thumbnails/45.jpg)
Bibliography III
Enrique G. Mendoza, Vincenzo Quadrini, and Jose-Vıctor
Rıos-Rull. Financial integration, financial development, and
global imbalances. Journal of Political Economy, 117(3):
371–416, 2009. doi: 10.1086/599706. URL
https://doi.org/10.1086/599706.