Banking and the Endogenous Money Supply as viewed from a

Download Report

Transcript Banking and the Endogenous Money Supply as viewed from a

AN INTEGRATED APPROACH TO THE STUDY OF
RESERVE EARNING ECONOMIES
ÁNGEL GARCÍA BANCHS
University of Siena,
Annual Meeting
27-06-2008, Siena, Italy
STRUCTURE OF THE PRESENTATION
1. Introduction (motivation and objectives).
2. Methodology: Stock-flow consistency approach to
macroeconomics.
3. The Neoclassical versus the Post Keynesian model.
4. Features of my model − exogenous and endogenous
variables.
5. Final remarks.
2
1.
INTRODUCTION (Motivation)
Interesting findings from previous work:
“International Monetary Asymmetries and the Central
Bank”, forthcoming in Revista Investigación Económica,
UNAM, Mexico, Jul-Sep 2008.
(García, Mata and Nell, 2008).
 The international monetary system is asymmetric.
3
1.
INTRODUCTION (Motivation)
Interesting findings from previous work:
 All national states can circulate domestically their
own currencies but not all of them can do so
internationally.
 The world has become divided among reserve issuing
and reserve earning economies.
 This transformation occurred after WWII, when the
elastic supply of US dollars arrived to replace the
scarce supply of gold as international means of
settlement.
4
1.
INTRODUCTION (Motivation)
Interesting findings from previous work:
 The quantity effect and the price effect.
 The former means reserve earning economies must
be concerned with the preservation of a minimum stock
of foreign currency assets, while the latter implies they
must be concerned with the stability of the foreign
exchange rate.
5
1.
INTRODUCTION (Motivation)
Interesting findings from previous work:
 A strong supply side connection between the
short-term rate of interest, the exchange rate, and the
stock of foreign reserve assets of the central bank.
Different from traditional demand-side link.
6
1.
INTRODUCTION (Motivation)
Reserve
Issuing
Economies
Local Currency
Reserves
Interest
Rate
Reserve
Earning
Economies
Foreign Currency
Reserves
FX
Rate
Local Currency
Reserves
Interest
Rate
7
1.
INTRODUCTION (Motivation)
Interesting findings from previous work:
 Monetary policy is more flexible but less influential
in reserve issuing economies, and less flexible but more
influential in reserve earning ones.
 The degree of interest rate exogeneity is much lower
in reserve earning economies than in reserve issuing
ones.
 Caveat: International monetary asymmetries affect
the behavior and balance sheet structure of the
institutional sectors in the economy.
8
1.
INTRODUCTION (Objective)
Simulate changes in parameters so as to compare the two
models: reserve earning and reserve issuing economies.
Two different papers.
Firms (F), Households (H), Commercial Banks (B), the
Central Bank (CB) the Government (G) and the Rest of
the World (ROW).
 How the rate of growth differs in the two economies.
 How stocks of wealth, flows of consumption and
investment and prices evolve over time.
9
2.
METHODOLOGY: STOCK-FLOW
CONSISTENCY APPROACH
Two views but same methodology:
The New Haven School, led by James Tobin at Yale
University, in the US, and
The Cambridge School, led by Wynne Godley, in the UK
(Godley and Lavoie, 2007).
The methodology is exactly the same. But the behavior of
their models differ. The former is an orthodox approach
and the latter is a Post Keynesian heterodox approach.
(What variables are exog/endog?)
10
2.
METHODOLOGY: STOCK-FLOW
CONSISTENCY APPROACH (Cont)
 Sectoral budget and system wide constraints:
“The fact that money stocks and flows must satisfy
accounting identities in individual budgets and in an
economy as a whole provides a fundamental law of
macroeconomics analogous to the principle of
conservation of energy in physics”.
Godley and Cripps (1983: p. 18)
11
2.
METHODOLOGY: STOCK-FLOW
CONSISTENCY APPROACH (Cont)
Stock-flow norms which may be self-imposed or
inflicted by other institutional sectors.
Appropriate use of lagged dynamics to make sure
causes precede effects.
Several assets and rates of returns both, private and
public, and short and long (Brainard and Tobin, 1968).
Financial and monetary policy is considered.
12
2.
METHODOLOGY: STOCK-FLOW
CONSISTENCY APPROACH (Cont)
 Sectoral Budget constraint (Vertical):
Net WealthI  Net Regular IncomeI  CapitalGainsI
 System wide constraint (Horizontal N-1 sectors):
I
Net Wealth in the whole economy   Net Wealth
i
i 1
 Balance sheet, revaluation and
transactions matrices.
13
2.1 Balance Sheet Matrix of a REE
F
Real Assets
Inventories
Fixed Capital
Local Currency
Financial Assets
Cash
Bank Reserves
Government Dep. in CB
Current Account Deposits
Saving Account Deposits
Government Bills
Central Bank Bills, CDs
Bank Bills, CDs
Firm Bills, Comm. Paper
Government Bonds
Corporate (Firm) Bonds
Credit Loans
Lender of Last Resort
Firm Equities
H
S&D
S&D
D
D
D
D
D
D
D
D
D
D
S
D
S
D
D
D
D
D
D
S
D
B
CB
G
ROW
IN
K
D
D
D
S
S
D
D
S
D
D
D
S
D
S
S
D
D
D
S
S
Balance
S
D
S
S
D
0
0
0
0
0
0
0
0
0
0
0
0
0
0
14
2.1 Balance Sheet Matrix of a REE (Cont)
Foreign Currency
Financial Assets
Cash
Off-shore Deposits
Foreign Government Bills
Foreign Government Bonds
Foreign Corporate Bonds
Foreign Credit Loans
Foreign Equities
Domestic Bank Reserves
Domestic Deposits
Domestic (Sovereign)
Government Bonds
Domestic Corporate Bonds
Domestic Credit Loans
Balance
Sum
F
H
B
CB
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
G
S
S
S
S
S
S
S
D
D
D
D
D
S
D
D
D
S
D
VF
0
D
VH
0
S
VCB
0
ROW Balance
D
0
0
0
0
0
0
0
0
0
D
0
D
0
0
-(IN+K)
0
S
S
0
0
VG
0
Vrow
0
15
2.2 Revaluation Matrix of a REE
Real Assets
Fixed Capital
Local Currency
Financial Assets
Govt. Bonds
Firm Bonds
Firm Equities
Bank Capital
Foreign Currency
Financial Assets
Cash
Off-shore Deposits
Foreign Govt. Bills
Foreign Govt. Bonds
Foreign Firm Bonds
F
H
+
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
B
CB
G
ROW
+
+
-
0
0
0
0
+
-
+
+
+
Balance
-
0
0
0
0
0
16
2.2 Revaluation Matrix of a REE (Cont)
F
Foreign Currency
Financial Assets
(Continued)
Foreign Loans
Foreign Equities
Domestic Bank Reserves
Domestic Deposits
Domestic (Sovereign)
Government Bonds
Domestic Corporate Bonds
Domestic Loans
H
-
B
CB
G
-
+
+
+
+
+
-
+
+
-
Balance
+
+
0
0
0
0
+
0
+
0
0
-
+
ROW
17
2.3 Transactions Matrix of a REE
Domestic Economy
Firms
Households
Cur Cap Cur
Income
process
Consumption
Govt. Exp.
Fixed Invest.
Inventory Acc.
Exports
Imports
GDP
Income Tax
Wages
Inventory
Financing Cost
Entrepreneurial
Profits
Bank Profits
CB Profits
+
+
+
+
+
-Y
-
Cur
CB
Gov
Cap Cur Cap Cur Cap
-
-
+
Y
+
-
Cap
Banks
Row
F, H, B,
Balance
CB, G
Cur Cap
-
+
-
+
+
+
+
+
-
0
0
0
0
0
0
0
0
0
0
+
-
+
-
0
0
0
18
2.3 Transactions Matrix of a REE (Cont)
Domestic Economy
Row
F, H, B,
Firms
CB
Gov
Households Banks
CB, G Balance
Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap
Rent Transfers
Local Currency
SA Deposits
Govt. Bills
CB Bills, CDs
Bank Bills, CDs
Govt. Bonds
Credit Loans
LOR
Foreign Currency
Off-shore Deposits
Foreign Govt. Bills
+
+
+
+
+
+
+
+
-
+
+
+
+
-
+
-
0
0
0
0
0
0
0
-
+
+
+
+
+
-
0
+
+
+
+
-
0
19
2.3 Transactions Matrix of a REE (Cont)
Domestic Economy
Row
Househol
F, H, B,
Firms
Banks
CB
Gov
Balance
ds
CB, G
Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap
Rent Transfers
Foreign Currency
Foreign Govt. Bonds
Foreign Corporate Bonds
Foreign Credit Loans
Foreign Equities
Domestic Bank Reserves
+
+
+
Domestic Deposits
Domestic (Sovereign)
Government Bonds
+
+
+
+
-
+
+
+
-
0
0
0
0
0
+
0
+
0
-
-
20
2.3 Transactions Matrix of a REE (Cont)
Domestic Economy
Row
Househol
Gov
F, H, B,
Firms
Banks
CB
ds
CB, G
Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap
Stock Variations
Local Currency
Cash
Bank Reserves
Govt. Dep. in CB
CA Deposits
SA Deposits
Govt. Bills
CB Bills, CDs
Bank Bills, CDs
Corporate Bills
Govt. Bonds
Corporate Bonds
Credit Loans
Loan Defaults
Lender of Last R.
Firm Equities
-
-
-
-
+
+
+
+
+
+
-
+
+
+
+
+
+
+
-
+
+
-
+
-
-
-
Balance
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
21
2.3 Transactions Matrix of a REE (Cont)
Firms
Domestic Economy
Househol
Banks
CB
ds
Gov
Row
F, H, B,
CB, Gov Balance
Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap Cur Cap
Foreign Currency
Cash
Deposits abroad
Foreign Govt. Bills
Foreign Govt. Bonds
Foreign Firm Bonds
Foreign Loans
Foreign Equities
Domestic Bank Reserves
Domestic Dep.
Domestic Govt. Bonds
Domestic Firm Bonds
Domestic Loans
Loan Defaults
Sum
-Δ
-Δ
-Δ
-Δ
-Δ
-Δ
-Δ
-Δ
-Δ
+Δ
-Δ
-Δ
-Δ
+Δ
+Δ
+Δ
+Δ
+Δ
-Δ
+Δ
-Δ
-Δ
-Δ
+Δ
-Δ
-Δ
-Δ
-Δ
-Δ
+Δ
+Δ
+Δ
0
-Δ
0
-Δ
+Δ
+Δ
+Δ
-Δ
-Δ
-Δ
-Δ
-Δ
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
22
3.
THE NEOCLASSICAL MODEL VERSUS
THE POST KEYNESIAN MODEL
Both share the same methodology, but:
 In the Neoclassical model (NCM) agents maximize
utility and profits. There is need and room for the
rational expectations hypothesis.
 In the Post Keynesian model (PKM), procedural
rationality and adjustment to disequilibrium is assumed.
23
3.
THE NEOCLASSICAL MODEL VERSUS
THE POST KEYNESIAN MODEL (Cont)
 In the NCM: perfect information assumption.
 In the Post Keynesian model (PKM): generic
uncertainty, liquidity preference, norms and targets
which determine behavior.
 For instance: inventories to sales ratio, bills and
reserves to deposits, income to wealth, foreign reserves
to imports, debt to GDP, foreign debt to exports,
etc…although Tobin and some Neoclassical Keynesians
have also used stock-flow norms.
24
3.
THE NEOCLASSICAL MODEL VERSUS
THE POST KEYNESIAN MODEL (Cont)
The particular shape of expectations is not crucial, as
any mistaken expectations lead to unexpected variations
in inventories, money and wealth, signaling the need for
a change in behavior.
Other features of the PKM: the principle of effective
demand, imperfect competition, mark-up pricing, fixed
technical coefficients, conflictive income distribution, the
role of capacity utilization and retained profits, among
many others
25
3.
THE NEOCLASSICAL MODEL VERSUS
THE POST KEYNESIAN MODEL (Cont)
 But above all, the fact that money is endogenous.
 For any good or asset, if quantities are endogenous,
prices must be exogenous, and the converse.
Thus, if money is endogenous, the interest rate must be
exogenous.
 This will never be accepted by neoclassical economists,
as their theory of prices and distribution would collapse.
26
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
 My model incorporates our previous findings, namely
that international monetary asymmetries are largely
responsible for determining the behavior and balance
sheet structure of the institutional sectors in the
economy.
 327 equations. Why? The old answer is also the best
one: the real world is complex, and computational
power makes non-analytical results from simulations
possible and amenable.
27
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Firms (125 equations – 3 dummy variables).
Production and Real Investment Decisions:
Capital, intermediate and final goods are produced
and imported, locally consumed and exported, and it is
assumed there are inventories only in the latter case.
Inventories act as a buffer stock (asset side).
Intermediate goods are required in accordance to fix
technical coefficients.
28
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Production and Real Investment Decisions:
Investment in capital goods depends on an exogenous
component and on capacity utilization.
 Both “ig” and “kg” are locally produced and imported
in some fixed proportions that are assumed to depend on
the structure and degree of development of the economy,
and not on relative prices.
29
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Costs of Production:
Inflation is a conflictive-claims process.
Workers aim at a real wage rate-target whose size
varies continuously with average trend productivity and
discontinuously with aggregate demand (e.g. with the
employment rate), a sort of discontinuous Phillips curve
with an inelastic segment. For simplicity, productivity is
assumed to grow at an exogenous rate.
30
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Costs of Production:
 The desired level of employment depends on average
productivity and output. Yet, actual employment adjusts
only partially towards target.
 Unit costs of sales depend on the value of the wage bill
and imports bill. Historic unit costs apply on in the case
of final goods, as only those goods are accumulated in
inventories.
31
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Pricing:
 There is a total sales price, domestic price, export
price and imports price for every good. Total sales prices
are a mark-up over historic unit costs. Two alternative
cases: Argentina, China?.
 The ideal mark-up is that which would hypothetically
generate the exact amount of profits required by firms
to satisfy target retained earnings and dividend
payoffs when realized profits are equal to
planned profits.
32
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Financial Considerations:
 Most of the investment expenditure is financed by
profits, but the remaining part is financed by new issues
of local and foreign currency debt and equities.
Firms structure their portfolio of assets and liabilities
in accordance to interest rates and degrees of liquidity
preference in local and foreign currency. Rates paid by
corporate sector are market determined (endogenous).
Credit acts as a buffer stock (liability side).
33
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Households (48 equations).
Consumption and Real Investment Decisions:
 Modigliani consumption function with propensities to
consume out income (and consumer loans) and
(expected) wealth.
Investment in real estate and durables also depends on
income, loans and wealth.
34
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Financial Investment Decisions:
Households receive wage payments from firms, banks
and the government.
 They place their savings across diverse lc and fc
financial assets in accordance with their rates of interest
and liquidity preference in lc and fc.
35
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Financial Investment Decisions:
 In line with the CRL, credit to households is
constrained by income in a way which is inversely
related to the real interest rate (the burden of debt).
 Current account deposits are the buffer stock of
households.
36
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Banks (55 equations – 3 dummy variables).
Monetary and credit aggregates:
 Cash is held in fixed proportion to CA and SA
deposits.
 Bank reserves within the central bank are held in
fixed proportions to CA, SA and CDs.
 Banks also hold a fraction in secondary reserves
(PA bills).
37
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Monetary and credit aggregates:
 They cannot control directly the amount of bills they
hold, as the latter is a residual the central bank
accommodates once the demand for T-bills on the part
of the other sectors has been satisfied.
 Yet, they can indirectly influence the amount of bills
they hold: in the short-run resorting to the discount
window and in the long-run adjusting the rate they pay
on deposit certificates, increasing it when the ratio is
below target and reducing it when it is above.
38
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Monetary and credit aggregates:
 Banks accommodate holdings of (demand for)
deposits, setting/paying a mark-up rate above the rate
on PA bills.
 They also accommodate the demand for credit in lc
and fc on the part of F & H, setting the rate on loans as a
spread over the deposit rate. The spread depends on
CAR and profit targets.
Bills and (only temporarily) CB advances act as
buffer stocks (on the asset and liability sides).
39
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Government (36 equations – 5 dummy variables).
 Tax revenue (direct taxes paid by F, H, B, and indirect
taxes over fg).
 Nominal pure govt. expenditure: current (wage bill
and purchases of final goods) and capital expenditure in
infrastructure.
 Both are initially assumed to grow at exogenous rates.
40
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
 The primary deficit plus interest payments determine
the financial deficit.
Clearly, as money is endogenous and interest rates are
exogenous, fiscal monetary expansions arising from
government deficits must be absorbed: government may
increase its fraction of deposits at CB, the latter may
increase the legal rate of reserve requirements, and bills
and bonds must be issued. Otherwise r↓, GIR↓, xr↑.
Govt. bonds are supplied on demand. But bills are
supplied in accordance to cash flow requirements.
41
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
 Short-term rate, GIR and foreign currency debt:
FCD
r
FCD to GDP
or to exports
ratio
Flexible
Non flexible
Flexible
ΔGIR>0
ΔFCD>0
ΔGIR=0
ΔFCD=0
Inflation or loss of
competitiveness
Non flexible
T
rBG  rBG
 z8   BG
ΔGIR>0
ΔFCD<0
T
rBG  rBG
T
rBG  rBG
e.g. China
Public
Admin
Whole
Economy
e.g. China
GIR
Public
Admin
Whole
Economy
GIR
42
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
 Foreign reserves and foreign currency debt act as buffer
stocks.
 So doing the government is able to stabilize the xr and
preserve a minimum level of GIR. But all depends on
the possibility to place more fc debt.
Why would a RIE (like the EU or the US) issue debt in
fc? That would only be “una pazzia”.
43
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
Central Bank (36 equations – 5 dummy variables).
 Accommodates the overall demand for base money.
 Yet, it sets the rate on reserve requirement, adjusting
to absorb/sterilize foreign currency inflows (e.g. China).
 Holdings of
accommodated.
GD
within
the
CB
are
also
44
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
 CB makes effective short-term rate by managing the
supply of bills of the PA. Opposed to the case of RIEs,
this is mainly done on the liability side.
 The fact is the demand for T-Bills on the part of REE
CBs is rather small, sometimes even negligible due to
regulation or self-imposed restrictions.
 Thus, in REEs base money is created through the
increase in foreign reserves on the asset side. But it is
destroyed on both the asset side and the liability side
through reductions in foreign reserves and increases in
CB bills and GDs within CB.
45
4.
FEATURES OF MY MODEL − EXOG
VS ENDOG VARIABLES.
ROW (42 equations).
 Foreign currency assets supplied by the ROW
accommodate demand, except for the supply of bank
loans to subsidiaries which are assumed to grow at the
rate of exports. And, for simplicity, it is assumed
domestic fc deposits held by non residents grow at the
rate of imports.
 Foreign interest rates are treated exogenously and
domestic fc rates on government and corporate bonds
are market-determined.
46
4.
FINAL REMARKS.
 My next paper will study the much simpler case of
RIEs (like the US or the EU).
 I will run diverse simulations for this model and the
next model to compare the results in both artificial
economies.
How the rate of growth differs in the two economies
under different conditions (fc inflows and outflows)?
 How stocks of wealth, flows of consumption and
investment and prices evolve over time, and
many, many, other.
47
APPENDIX
48
Table 1. The Balance Sheet Matrix of a REE
Relations with the
Rest of the World
Domestic Economy
Firms
Households
Banks
CB
Gov
Balance
F, H, B, CB, Gov
Real Assets
Inventories
IN
IN
Fixed Capital
KF
KH
CashF
CashH
KG
K
CashG
GDG
0
0
0
M1G
0
Local Currency
Financial Assets
Cash
Bank Reserves
Government Dep. in CB
CashB
BResB
Current Account Deposits
M1F
M1H
-M1
Saving Account Deposits
Government Bills
Central Bank Bills, CDs
Bank Bills, CDs
Firm Bills, Comm. Paper
Government Bonds
M2F
BGF
M2H
BGH
BBF
-BF
BLGF*pblG
BBH
BFH
BLGH*pblG
-M2
BGB
BCBB
-BB
BFB
BLGB*pblG
-BLF*pblF
-LFB
BLFH*pblF
-LHB
Corporate (Firm) Bonds
Credit Loans
Lender of Last Resort
Firm Equities
Bank Capital
- eF*peF
eFH*peF
OFBH
BLFB*pblF
LB
-LB
-CashCB
-BResCB
-GDCB
BGCB
-BCB
0
0
0
0
0
0
-BG
-BLG*pblG
LCB
eFrow*peF
- OFBH
0
0
0
0
0
49
Table 1. The Balance Sheet Matrix of a REE (Continued)
Relations with the
Rest of the World
Domestic Economy
Firms
Households
Banks
CB
Cash
Off-shore Deposits
xr*$CashF
xr*$MrowF
xr*$CashH
xr*$MrowH
xr*$CashB
xr*$MrowB
Foreign Government Bills
xr*$BG-rowF
xr*$BG-rowH
xr*$BG-rowB
xr*$CashCB
xr*$MrowCB
xr*$BG-
xr*$BLG-
xr*$BLG-
xr*$BLG-
Gov
Balance
F, H, B, CB, Gov
Foreign Currency
Financial Assets
Foreign Government Bonds
row
F
row
*$pblG-row
Foreign Corporate (Firm)
Bonds
Foreign Credit Loans
H
*$pblG-row
xr*$BLFrow
row
row
CB
B
*$pblG-row
H
*$pblF-row
-xr*$LF
row
-xr*$LBrow
xr*$erowH
*$perow
Foreign Equities
Domestic Bank Reserves
Domestic Deposits
xr*$MBF
xr*$MBH
xr*$BRes
-xr*$MB
Domestic (Sovereign)
Government Bonds
xr*$BLGF
*$pblG
xr*$BLGH
*$pblG
xr*$BLGB
*$pblG
Domestic Corporate (Firm)
Bonds
-xr*$BLF
*$pblF
xr*$BLFH
*$pblF
Domestic Credit Loans
Balance
Sum
-xr*$LFB
VF
0
VH
0
xr*$LB
0
0
-xr*$Cashrow
-xr*$Mrow
0
0
-xr*$BG-row
0
-xr*$BLG-row
*$pblG-row
0
-xr*$BLF-row
*$pblF-row
0
xr*$Lrow
0
-xr*$erow
*$perow
0
-xr*$BRes
xr*$MBrow
-xr*$BLG
*$pblG
0
0
xr*$BLGrow
*$pblG
0
xr*$BLFrow
*$pblF
0
0
-(IN+K)
0
VCB
VG
Vrow
0
0
0
50
Table 2. Revaluation Matrix of a REE
Relations with the
Rest of the World
Domestic Economy
Firms
Households
Δp*k-
Δp*k-1H
Government Bonds
ΔpblG*BLG-
ΔpblG*BLG-1H
ΔpblG*BLG-1B
Firm Bonds
-ΔpblF*BLF-
ΔpblF*BLF-1H
ΔpblF*BLF-1B
-ΔpeF*eF-1
ΔpeF*eF-1H
{ΔOFBH}
Real Assets
Fixed Capital
Local Currency
Financial Assets
Firm Equities
Bank Capital
Banks
{-ΔOFB
CB
Gov
Balance
F, H, B, CB, Gov
Δp*k-1G
Δp*k-1
-ΔpblG*BLG-1
0
0
ΔpeF*eF-1row
0
0
H}
Foreign Currency
Financial Assets
Cash
Δxr*$Cash-
Δxr*$Cash-1H
Δxr*$Cash-1B
Δxr*$Cash-1CB
-Δxr*$Cashrow-1
0
Off-shore Deposits
Δxr*$Mrow-
Δxr*$Mrow-1H
Δxr*$Mrow-1B
Δxr*$Mrow-1CB
-Δxr*$Mrow-1
0
Foreign Gov Bills
Δxr*$Brow-
Δxr*$Brow-1H
Δxr*$Brow-1B
Δxr*$Brow-1CB
-Δxr*$Brow-1
0
Foreign Gov
Bonds
Foreign Firm
Bonds
$BLG−row$BLG−row-1H
*
*
[(Δxr*$pblG−row) [(Δxr*$pblG−row)
+
+
(Δ$pblG−row*xr-1)] (Δ$pblG−row*xr-1)]
$BLF−row-1H
*
[(Δxr*$pblF−row)
+
(Δ$pblF−row*xr-1)]
-$BLG−row-1
*
[(Δxr*$pblG−row)
+
(Δ$pblG−row*xr-1)]
-$BLF−row-1
*
[(Δxr*$pblF−row)
+
(Δ$pblF−row*xr-1)]
0
0
51
Table 2. Revaluation Matrix of a REE (Continued)
Relations with the
Rest of the World
Domestic Economy
Firms
Households
Banks
CB
Gov
Balance
F, H, B, CB, Gov
Foreign
Currency
Financial Assets
(Continued)
Foreign Loans
-Δxr*$LF-1row
-Δxr*$LB-1row
H
$erow-1
*
[(Δxr*$perow)
+
(Δ$perow*xr-1)]
Foreign Equities
Domestic
Bank Reserves
Domestic
Deposits
Δxr*$BRes-1
Δxr*$MB-
Δxr*$MB-1H
Domestic
(Sovereign)
Government
Bonds
$BLG*
[(Δxr*$pblG)
+
(Δ$pblG*xr-1)]
$BLG-1H
*
[(Δxr*$pblG)
+
(Δ$pblG*xr-1)]
Domestic
Corporate
(Firm) Bonds
-$BLF-1
*
[(Δxr*$pblF)
+
(Δ$pblF*xr-1)]
$BLF-1H
*
[(Δxr*$pblF)
+
(Δ$pblF*xr-1)]
Domestic Loans
-Δxr*$LF-1B
0
-$erow-1
*
[(Δxr*$perow)
+
(Δ$perow*xr-1)]
0
-Δxr*$BRes-1
0
-Δxr*$MB-1
-$BLG-1
*
[(Δxr*$pblG)
+
(Δ$pblG*xr1)]
Δxr*$L-1B
Δxr*$L-1row
Δxr*$MB, -1row
0
$BLG-1row
*
[(Δxr*$pblG)
+
(Δ$pblG*xr-1)]
0
$BLF-1row
*
[(Δxr*$pblF)
+
(Δ$pblF*xr-1)]
0
0
52
Table 3. Transaction Matrix of a REE
Firms
Cur
Cap
Income process
Consumption
Gov
Expenditures
Fixed Investment
Inventory Acc.
Exports
Imports
GDP
Income Tax
Wages
Inventory
Financing Cost
Entrepreneurial
Profits
Bank Profits
CB Profits
C
Domestic Economy
Households
Banks
Cur
Cap
Cur
Cap
CB
Cur Cap
-C
-FF
-G
-IF
-ΔIN
-IH
0
-IG
-X
xr*IMC
+xr*IMIG
+xr*IMK
Y
-TH
WB
FUFF
FDFH
+rBF-1
*BF-1H
+BLF-1H
+xr*
$BLF-1H
FDBH
Balance
0
G
I
ΔIN
X
-xr*IMC
xr*IMIG
-xr*IMK
-Y
-TF
-WBF
-rL-1
*IN-1
Gov
Cur
Cap
Row
F, H, B, CB, Gov
Cur
Cap
-TB
0
0
0
0
T
-WBG
rL-1
*IN-1
rBF-1*BF-1B
+BLF-1B
+rL-1*[LF-1B
-NPL- IN-1]
-rM-1*M2-rBB-1*BB+xr*
[$rL-1B*$LF-1B
-$rM-1B*$MB-1F]
-FB
FUBB
0
0
0
0
-rBG-1
*BG-BLG-xr*
$BLG-
-FCB
FCB
G
xr*
[$BLF-1row
-$rM-1row
*$Mrow-$rBG,row-1
*$BG,row-$BLG,row-1F]
0
0
0
53
Table 3. Transaction Matrix of a REE (Continued)
Domestic Economy
Households
Firms
Cur
Rent Transfers
Local Currency
Saving Account
Deposits
Cur
rM-1
*M2-1H
rBG-1
*BG-1H
Cap
Cur
CB
Cap
Cur
Bank Bills, CDs
rBB-1
*BB-
rBB-1
*BB-1H
-rM-1
*M2-1
rBG-1
*BG-1B
rBCB-1
*BCB-1B
-rBB-1
*BB-1
Government
Bonds
BLG-
BLG-1H
BLG-1B
-rL-1
*LH-1
rL-1
*LH-1
-rLCB-1
*L-1
rLCB-1
*L-1CB
xr
*$rM*-1
*$Mrow-1B
xr
*$rB*-1
*$Brow-1B
xr
*$rM*-1
*$Mrow-1CB
xr
*$rB*-1
*$Brow-1CB
Government Bills
rM-1
*M2rBG-1
*BG-
Cap
Banks
Central Bank
Bills, CDs
Credit Loans
Lender of Last
Resort
Foreign
Currency
Off-shore
Deposits
Foreign
Government Bills
xr
*$rM*-1
*$Mrowxr
*$rB*-1
*$Brow-
xr
*$rM*-1
*$Mrow-1H
xr
*$rB*-1
*$Brow-1H
Gov
Cap
Cur
Cap
Row
F, H, B, CB,
Gov
Cur
Cap
Balance
0
rBG-1
*BG-1CB
-rBCB-1
*BCB-1
-rBG-1
*BG-1
0
0
0
-BLG-1
0
0
0
-xr
*$rM*-1
*$Mrow-1
-xr
*$rB*-1
*$Brow-1
0
0
54
Table 3. Transaction Matrix of a REE (Continued)
Domestic Economy
Households
Banks
Firms
Cur
Rent Transfers
Foreign
Currency
Foreign
Government
Bonds
Foreign
Corporate
(Firm) Bonds
xr
*$BLG-row-
Foreign Credit
Loans
-xr
*$rL*-1
*$LF-1row
Cap
Cur
Cap
Cur
Cap
CB
Cur
Gov
Cap
Cur
-xr
*$BLG-row-
xr
*$BLG-row-1H
-xr
*$BLF-row-1
-xr
*$rL*-1
*$LB-1row
xr
*$rL*-1
*$LB-1row
-xr
*$FDF-row
xr
*$FDF-rowH
xr
*$rB*-1
$BRes
Domestic Bank
Reserves
Domestic
Deposits
xr
*$rM-1
*$M-
xr
*$rM-1
*$M-1H
Domestic
(Sovereign)
Government
Bonds
xr
*$BLG-
xr
*$BLG-1H
0
1
xr
*$BLF-row-1H
Foreign Equities
Cap
Row
F, H, B, CB,
Balance
Gov
Cur
Cap
-xr
*$rB*-
0
0
0
0
1
$BRes
-xr
*$rM-1
*$M-1
-xr
*$BLG-1
xr
*$rM-1
*$M-1row
0
xr
*$BLG-1row
0
55
Table 3. Transaction Matrix of a REE (Continued)
Firms
Cur
Cap
Stock Variations
Local Currency
Cash
Bank Reserves
Govt. Dep. in CB
Current Account
Deposits
Saving Account
Deposits
Govt. Bills
Central Bank
Bills, CDs
Bank Bills, CDs
Firm Bills,
Comm. Paper
Govt. Bonds
Corporate (Firm)
Bonds
Credit Loans
Loan Defaults
Lender of Last R.
Firm Equities
-ΔCashF
Households
Cur
Cap
-ΔCashH
Domestic Economy
Banks
CB
Cur
Cap
Cur
Cap
-ΔCashB
-ΔBResB
ΔCashCB
ΔBResCB
ΔGDCB
-ΔM1F
-ΔM1H
ΔM1
-ΔM2F
-ΔM2H
ΔM2
-ΔBGF
-ΔBGH
-ΔBGB
-ΔBGCB
-ΔBCBB
ΔBCB
-ΔBBF
ΔBF
÷(1+ΩbF)
-ΔBLGF
*pblG
ΔBLF
*pblG
÷(1+ΩblF)
ΔLFB
NPL
ΔeF*peF
-ΔBBH
-ΔBFH
÷(1+ΩbF)
-ΔBLGH
*pblG
-ΔBLFH
*pblG
÷(1+ΩblF)
ΔLHB
-ΔeFH*peF
ΔBB
-ΔBFB
÷(1+ΩbF)
-ΔBLGB
*pblG
-ΔBLFB
*pblG
÷(1+ΩblF)
-ΔLB
-NPL
ΔLB
Gov
Cur
Cap
Row
F, H, B, CB, Gov Balance
Cur
Cap
-ΔCashG
-ΔGDG
0
0
0
-ΔM1G
0
0
ΔBG
0
0
0
0
ΔBLG
*pblG
-ΔBLGrow
*pblG
-ΔBLFrow
*pblG
÷(1+ΩblF)
0
0
0
0
0
-ΔLCB
-ΔeFrow
*peF
0
56
Table 3. Transaction Matrix of a REE (Continued)
Firms
Cur
Cap
Domestic Economy
Households
Banks
Cur
Cap
Cur
Cap
Cur
CB
Cap
Gov
Cur
Cap
Row
F, H, B, CB, Gov
Cur
Cap
Balance
Foreign Curr.
Cash
Deposits
abroad
Foreign Govt.
Bills
Foreign Govt.
Bonds
xr*Δ$CashF
-xr*Δ$CashH
-xr*Δ$MrowF
-xr*Δ$MrowH
-xr
*Δ$BrowF
-xr
*Δ$BLG-rowF
*$pblG-row
-xr
*Δ$BrowH
-xr
*Δ$BLG-rowH
*$pblG-row
-xr
*Δ$BLF-rowH
*$pblF-row
Foreign Firm
Bonds
Foreign Loans
Foreign
Equities
Domestic
Bank Reserves
Domestic Dep.
Domestic
Govt. Bonds
Domestic Firm
Bonds
Domestic
Loans
Loan Defaults
Sum
xr*Δ$LFrow
xr*Δ$Cash
xr*Δ$CashC
xr*Δ$Cashro
B
B
w
xr*Δ$MrowB
-xr
*Δ$BrowB
xr*Δ$MrowCB
-xr
*Δ$BrowCB
xr*Δ$Mrow
xr
*Δ$Brow
xr
*Δ$BLG-row
*$pblG-row
xr
*Δ$BLF-row
*$pblF-row
-xr*Δ$Lrow
xr*erow
*Δ$perow
xr*Δ$LBrow
H
-xr*erow
*Δ$perow
-xr*Δ$MBF
-xr*Δ$MBH
-xr*Δ$BLGF
*$pblG
-xr*Δ$BLGH
*$pblG
xr
*Δ$BLF
*$pblF
-xr
*Δ$BLFH
*$pblF
-xr*
Δ$BRes
xr*Δ$MB
xr*
Δ$BRes
0
0
0
0
0
0
0
0
xr
*Δ$BLG
*$pblG
-xr*Δ$MBrow
-xr
*Δ$BLGrow
*$pblG
-xr
*Δ$BLFrow
*$pblF
0
0
0
xr*Δ$LFB
-xr*Δ$LB
0
NPL
0
-NPL
0
0
0
0
0
0
0
57
58
59
60
61
62
63
64
65
66