Endogenous money, monetary circuit and demand-led growth
Sergio Cesaratto
Dipartimento di economia politica e statistica Università di Siena
Cesaratto@unisi.it
Background: output and growth
• I will be rather elementary: clever things can be expressed (and better understood) in simple terms.
• Refusal of Say’s Law that supply generates its own demand (neoclassical version there is a natural interest rate that equilibrates S and I)
• Refusal based on the result of the capital theory controversy factors’ demand function are not well behaved, a natural interest rate equilibrating S and I simply do not exist
• The level of output is determined by aggregate demand Keynes’ multiplier
• Growth of output is determined by growth of aggregate demand Hicks’/Ackley’s, Serrano’s supermultiplier
m Z g
v t
E c G
m C g
v t
Y c
en e a
n
) 1
( 1
) 1 ) (
1 ( 1
1
D e n
g Y v
I g e g
z
Background: money
• Endogenous money theory (EDMT)
• I shall argue that EDMT is a fact (not a theory), and therefore can be shared also by neoclassical (marginalist) economists
• It is indeed shared by economists at central banks besides heterodox
economists. Knut Wicksell, perhaps the greatest marginal economist believed in it (like Claudio Borio chief economist at BIS)
• Exogenous money view EXMT (textbooks) Reserves create deposits that
create loans money/deposit multiplier theory: deposits (loans) are a multiple of the monetary base created by the central bank.
Reserves deposits loans
• EDMT: banks create loans, loans generate deposits, reserves are created on demand by the CB.
Loans deposits reserves
Exogenous and endogenous view of money/credit
• Behind the EXMT there is the “loanable fund theory” banks intermediate saving, they lend deposits.
• As Jakab and Kumhof of Bank of England argue, the problem is where these deposits came from:
if from other banks, well one bank expands loans and another has to reduce them. They make fun of traditional theory, showing that these deposits must come from outside the banking system, that is they must consist of saved-corn (next slide)
• The EDMT argue that banks create loans:
• What about reserves? They are created by the CB on demand. If the CB does not satisfy the demand for reserves, banks will look for them in the wholesale interbank market and the target interest rate sets by the CB will levitate upward. (As we shall see the CB policy interest rate is the short one in the overnight monetary market).
EDMT and the S/I relation
• Suppose that the loan is used to buy an investment good, we have this sequence (E=equities, H= hoarding).
The entrepreneur buys the capital-good pre-financed by a bank (initial finance), and later fund it (final finance) issuing equities and long-term loans (returning the short-term loan to the bank). Only ex post S finances I.
An alternative view
• It is straightforward to observe that in any single period saving and investment are always equal (Dalziel 1996a p. 223; 1996b, p. 117). Take, for instance period 3: the income generated in this period (64 ua) is unspent until period 4 (it is “temporarily saved” in period 3, so to speak);
summed to proper saving of periods 2 and 3 we obtain 100 ua. From period 2, therefore, although changing hands, the deposit consists of proper saving.
Initial and final finance: production or final purchase?
• Keynes (1937a, p. 222): “investment market can become congested through a shortage of cash. It can never become congested through shortage of saving. This is the most fundamental of my
conclusions within this field.” For a similar stance by authoritative conventional economists see Borio e Dysiatat (2011, 2015).
• With regard to the purchase of a capital good, we may at this point distinguish between financing (initial finance) and funding (final finance).
• Two questions:
• In the case above, both initial and final finance concern the final purchase of a capital good.
According to Keynes, Augusto Graziani, Paul Davidson and others, initial finance regards financing production of the capital good in advance of demand, while final finance funds the final purchase.
• Second, in what sense conventional (marginal) economics can share this endogenous money view?
• We shall begin from the second question, but before that, let us dwell a moment upon he Monetary Circuit Theory which has been very popular among heterodox economists.
Digression: Monetary Circuit Theory (Parguez, Graziani, Lavoie, Nell..)
• The MCT emphasizes the role of the financial sector in ignition of the economic circuit. The
emphasis is on the production side, that is, on the role of endogenous credit/money creation in financing production costs (mainly wage costs). The summary of the phases in the circuit drawn from is sufficiently representative:
• 1) Banks grant (…) the financing requested by firms, creating money (opening of the circuit);
• 2) Once financing has been obtained, firms buy inputs. Considering firms in the aggregate, their only expenditure coincides with the total wage bill; at this point money passes from firms to workers;
• 3) Once labour services have been purchased, firms carry out production (…)
• 4) At the end of the production process, firms put the goods on the market. It can be envisaged that firms set the sale price following a mark-up principle. Supposing workers have a propensity to consume equal to one, firms recover the entire wage bill and maintain ownership of a proportion (corresponding to the mark-up) of the goods produced. If the propensity to consume is less than one, … [workers] must make a further choice about how to use their savings, either hoarding (increase in cash reserves) or investing (purchase of shares). If all money savings are invested in shares on the financial market, firms manage to recover the whole wage bill;
• 5) Once goods and shares have been sold, firms repay the banks (closure of the circuit).
Monetary Circuit Theory: problems
• Main problem: but if demand consists of of wages only, where are profits? I will suggest a solution.
• Additional problem: production decisions are not explained using the Theory of
effective demand (The Treatise rather than the General Theory is the Mecca of
circuitists)
Knut Wicksell (and the red fish)
• A standard marginal economist (unfortunately, especially if s/he is young) has been compared to a red fish (short memory, happy in her bowlthe red wish doesn’t learn history of economic thought, s/he is a sort of creationist). Modern marginal theory is a continuous re-discovery of Wicksell (never fully understanding him, see Woodford). (KW was aware that with heterogenous capital goods marginal theory is flawed, this is neglected).
• At university my maestro (the favourite pupil of Sraffa) taught marginal theory directly on Wicksell.
• So we learnt that: “[W]hen there is a developed banking system… the volume of bank loans is independent of the flow of money savings: ‘By the concentration in their hands of private cash holdings… [the banks] possess a fund for loans which is elastic and, on certain assumptions,
inexhaustible’ (Wicksell 1935, p. 194). Hence the banks can accommodate any variation in the demand for loans without changing their rates of interest and can thus severe the link between the market rate of interest and the ‘natural’ rate” (Garegnani, 1983, pp.44-45).
• In a Wicksellian context, the Keynesian proposition that it is saving that generates investment can thus be re-proposed: monetary policy must target a monetary interest rate equal to the natural rate; at this rate, banks will finance through endogenous money creation a full employment level of investment that, in turn, generates a corresponding amount of full-employment saving. This must remind us that endogenous money, but also the Keynesian multiplier are facts, not concepts dependent on a specific theory.
Initial finance production financing Final finance purchase funding
An investment bankis engaged by an investing firm to fund (or long- term/final finance) the purchase of the capital good that costs 100 ua.
The order for the investment-good generates a corresponding production decision - so the first move is on the demand-side and not on the supply- side, as in MCT. Production takes time, and wage-costs to produce the capital-good, say 80 ua, are pre-financed by endogenous-money creation by a commercial bank (initial finance). Let us assume that workers consume all wages and capitalist save all profits. Assuming a wage-share of 0.8 (both in the investment and consumption goods sectors), the average marginal propensity to consume will also be 0.8. With these hypothesis, initial spending out of wage-payment (80 ua) will be 80 ua.
The newly generated income is 480 ua and saving 80 ua, equal to the initial spending. These savings are collected by the investment bank - referring to table 2, the collection may take the form of an issue of long- term assets [E] or collection of loans from banks that convert deposit [H]
in longer term loans. An investment bank does not create deposits, but intermediates savings; it can be a branch of a commercial bank,
transforming short-term deposits in long-term loans, or an independent firm that collect savings by issuing long-term bonds. Final income is equal to the wages anticipated in the investment sector (80 ua) in period 1, plus the income generated in the consumption sector in the subsequent periods (400 ua); at this stage saving are generated in the consumption sector only.
The question is that, at this stage (round 1), funding (80 ua) fall short of the cost of the capital good (100 ua). This demand-led monetary circuit may be closed by the investment bank borrowing the missing 20 ua from the commercial bank (round 2), to be lent to the investing firm that can thus fully regulate the payment for the capital good. Finally, in the hypothesis that the capital-good producer saves all her profits (20ua), the investment bank can collect them, returning the 20 ua loan to the commercial bank. Not surprisingly, final (R1+R2) income, consumption and saving are the same of table 1.
Extension I: initial finance concern all production
• The demand-led monetary circuit is based on the idea that, particularly where physical commodities are produced, production must anticipate actual sales. In the preceding examples, however, this happens only in the investment sector, while production in the consumption sector follow, like De Gaulle’s intendance (“L’intendance suivra!”). By extension, also production prepared in advance to meet induced consumption generated by the income multiplier process, is pre- financed. See Graziani’s (1984, p. 22) discussion and references to Keynes’
insistence that initial finance concern all production; the question is examined in Cesaratto 2017, section 4.4.
• Borio & Dysiatat also distinguish between initial finance (production) and final
finance (purchases): again, this is consistent with marginal theory.
Extensions II: autonomous components of AD
• As seen above, in the (super)multiplier analysis, the autonomous components/non capacity creating of AD govern the level and growth of output.
• By definition, these components are financed out of purchasing power creation by banks. It is a marginalist fantasy that:
• Saving from thrifty households finances (ex ante) Ca of profligate households (true only ex post)
• The government must collect taxes and saving before spending (it spends first, ex post collects tax revenues and saving)
• Saving from thrifty (CA>0) countries finances (ex ante) spending of profligate CA<0) countries (true only ex post)
• The story of Ca is quite similar to the investment story, so let us go to the two other more intriguing questions.
m Z g
v t
E c G
m C g
v t
Y c
en e a
n
) 1
( 1
) 1 ) (
1 ( 1
1
The State spends first
• That government spending generates – given sufficient spare capacity – tax revenues and saving (that finances deficit sending) is common knowledge. So the question is: how can the
government spend first?
• The Modern Monetary Theory (Chartalist) answer is that the State spends using its account at the CB. But CB financing of State spending is forbidden by the regulations.
• So called post-Chartalis view is that the government is financed by endogenous money creation by commercial banks as private units. This is a bit complicated and should be explored furtherly.
The State spends first: preliminary
• In the example of table 4 (adapted from Cesaratto 2016, table 3), let us assume that government spending is financed by issuing treasury bonds bought by the central bank (CB) or by commercial banks that will create purchasing power (a deposit) in favour of the Treasury (we will shortly expand this point). In a closed economy with spare capacity, given a saving propensity c = 0.7 and a tax rate t = 0.3, government spending of 100 ua (the purchase of an aircraft) would generate an additional income of 196 ua, fiscal revenues of 58.8 ua and additional saving of 41.2 ua. Therefore, 58.8 ua of the initial Treasury debt will not be rolled over, while households’ saving will “fund” the remaining 41.2 ua of government debt, for instance, by buying the treasury bonds from the CB or from the commercial banks. The point we bring home from table 4 is that government spending is financed by purchasing power creation (initial finance) and that this leads to ex post funding (final finance) by taxes and households’ saving.
• Let us then figure out how government can have access to purchasing poser creation.
Post-Chartalist view of the State spends first
In this example government spending is totally financed by the issue of T-bonds (is totally deficit spending). However, government spending generates both additional saving and tax revenues.
International capital flows
• Marginal view (Bernanke): International capital flows consist of saving lent by capital rich countries to capital poor partners - international loanable fund theory
• Alternative view (see also Borio and Dysiatat 2011, 2015): excess spending in CA deficit countries is financed (initial finance) by endogenous money creation by banks (these may be located in the deficit country, or in the core country or even in third countries).
• Funding of the CA (final finance) takes (typically) place (ex post) through interbank lending (this is a saving flow).
• The example below concern what happened in the EMU.
• The process begins:
The upper capital flow concerns financing (“pure monetary flow”), the lower capital flow regards funding (saving flow)
Capital gains as autonomous consumption (sort of)
• Jakab and Kumhof (2015, p. 12):“many modern banking systems, loans to finance investment in the real economy have become a fairly small part of overall bank lending, with another part financing consumption, and a third and much larger part financing the exchange of existing real or financial assets between different agents.” Banks creation of new purchasing power in support of the purchase of real or financial asset is at the basis of speculative bubbles in the price of these assets.
• This may have in turn two real effects in the economy: a direct effect when the sellers of the inflated assets realise their capital gains, that can then be consumed (or invested) sustaining
aggregate demand; and an indirect outcome since the increase in financial and real estate wealth my raise the marginal propensity to consume. While this is the traditional “wealth effect”, let us focus here on the former, direct channel.
Steindl’s plot of land
• Steindl provides an example of a plot of land (but it could well be another real-estate or financial asset) whose value has been rising (and is expected to do so). This expectation motivates the purchase of this “plot of land” by a buyer financed by bank credit that, in line with the former exposition, we may classify as initial finance. The seller “will use the proceeds of his sale in order to pay back the credit he had taken when he in turn bought the land”, being left with “a surplus, his realised capital gain” (ibid, p. 437). The banking system is left with an expansion of credit: the difference between the loan to the buyer of the land (efflux), and the reimbursement by the
seller (reflux).
• Capital gains may be unspent, Steindl argues. If they are spent, consumed or invested, there will be a positive effect on demand
• Capital gains are thus a form of autonomous spending financed by endogenous credit/money.
The difference between autonomous consumption and consumption out of capital gains is that in the former case the spender has a debt with a bank, whereas in the latter the capital gainer has not a debt with a bank, but the buyer of the “plot of land” has it. So, in final, in both cases
consumption is financed out of some household’s debt.
Initial finance production financing Final finance purchase funding
• It is too complicated to show now that also in the case of State and foreign spending, initial finance may concern production (endeavoured either by the expectation of State spending or of foreign demand)
• I just want to add that endogenous money creation is also behind financial bubbles and capital gains. The latter affect aggregate demand either indirectly through wealth effects, or directly when capital gains are spent.
• Steindl capital gains financed by endogenous money creation are a sort of autonomous consumption.
• A final result is that endogenous money creation sustain debt-financed spending that in the long run may reveal unsustainable. But this is capitalism.
• What we have argued so far can be shared also by open-minded marginal economists. The
difference is that they believe that as long as interest rates are at their natural level debt-spending is not destabilising. Crises would depend on Wicksellian mistakes in monetary policy.
• Those who reject the notion of natural interest rate may think of other debt-stabilisation rules (based on the relationship between nominal interest rates and nominal growth of GDB).
Summing up
• The paper is a contribution to the integration of the long-run theory of effective demand with elements from the monetary circuit and endogenous finance analyses.
• Autonomous demand, investment and capital gains are sustained by endogenous finance.
• The paper explored these mechanisms also in view of Keynes’ distinction between initial and final finance, vigorously backed by Graziani as well as by more orthodox authors such as Borio and Dysiatat.
• More specifically, the paper proposed an integration between the traditional Keynesian role of initial finance as supporting final demand (the ‘received view’) with the MCT emphasis of the role of initial finance as sustaining production decisions.
• In this respect we proposed a “demand-led monetary circuit theory” in place of the traditional production-led monetary circuit theory.
• Open-minded marginal economists may share aspects of this view, the difference is that they believe that as long as interest rates are at their natural level, debt-spending is not destabilising. Those who reject the notion of natural interest rate may think of other debt- stabilisation rules (e.g. based on the relationship between nominal interest rates and
nominal output growth).
References
• McLeay M., Amar, R. Ryland, T. (2014): ‘Money creation in the modern economy’, Bank of England, Quarterly Bulletin, No. 1, pp. 1-14.
• Jakab, Z., Kumhof, M. (2015): ‘Banks are not intermediaries of loanable funds - and why this matters’, Bank of England, Working Paper No. 529.
• Lavoie, M. “A Primer in Endogenous Credit-Money.” In L.P. Rochon, and S. Rossi (eds.), Modern Theories of Money. The Nature and Role of Money in Capitalist Economies, Cheltenham: Edward Elgar, 2005, pp. 506-543.
Working paper version (2002) available at:
http://aix1.uottawa.ca/~robinson/Lavoie/Courses/2007_ECO6183/childguide4.pdf)
• Graziani A. (1990): ‘The theory of the monetary circuit, Économies et Sociétés, 24(6), pp. 7-36.
• Lavoie, M. “The Monetary and Fiscal Nexus of Neo-Chartalism: A Friendly Critical Look.” Journal of Economic Issues, 42, 2013: 1-32. Working paper version (2011) available at:
http://www.boeckler.de/pdf/v_2011_10_27_lavoie.pdf
• Cesaratto, S. (2017) Initial and final finance in the monetary circuit and the theory of Effective Demand. 2016, Metroeconomica, Volume 68, Issue 2, May , Pages: 228–258.
• Cesaratto, S. (2016) The State Spends First: Logic, Facts, Fictions, Open Questions, Journal of Post Keynesian Economics, 39 (1), 44-71.
• Cesaratto, S. (2017), Beyond traditional monetary circuit: endogenous money, finance and the theory of long- period effective demand, Storep conference 2017, Piacenza, June.
