Keynesian Uncertainty, Money and Explicit Money Contracts
Individuals in the real world must decide whether past experience provides a
useful guide to the future. Should one presume that economic processes are
uniform and consistent so that events are determined, either by ergodic
stochastic processes or at least by specified and completely ordered prospects?
Can the agent completely dismiss any fear of tragedy during the time
between choice and outcome? Does the agent believe he or she is ignorant
regarding the future? No rule can be specified in advance regarding how
individuals decide whether they are in an objective, a subjective, or a true
uncertainty environment. However, their perception will make a difference to
their behaviour.
Keynes laid great stress on the distinction between uncertainty and probability,
especially in relation to decisions involving the accumulation of wealth
and the possession of liquidity. The essence of his General Theory involves
liquidity preferences and animal spirits dominating real expenditure choices.
Money plays a unique role in ‘ruling the roost’ among all assets (Keynes,
1936, p. 223) and it is non-neutral in both the short and long run (Keynes,
1973a, pp. 408–11). These claims, as Keynes made clear in his 1937 restatement
(Keynes, 1973b, pp. 112, 114) of where he saw his general theory ‘most
clearly departing from previous theory’, rested on the clear distinction between
the ‘probability calculus’ and conditions of uncertainty when ‘there is
no scientific basis to form any calculable probability whatever. We simply do
not know.’
Liquidity and animal spirits are the driving forces behind Keynes’s analysis
of long-period underemployment equilibrium, even in a world of flexible
prices. Neither objective nor subjective probabilities suffice to understand the
role of non-neutral money and monetary policy in Keynes’s underemployment
equilibrium analysis. It is not surprising, therefore, that unemployment
still plagues most twenty-first-century economies, since most economists still
formulate policy guidelines which are only applicable to a limited domain
where agents choose ‘as if’ they had specific and completely ordered knowledge
about the future outcomes of their actions.
In Davidson (1978, 1982), I have shown that the existence of the societal
institution of legally enforceable forward contracts denominated in nominal
(not real!) terms creates a monetary environment that is not neutral, even in
the long run. These legal arrangements permit agents to protect themselves to
some extent against the unpredictable consequences of current decisions to
commit real resources towards production and investment activities of long
duration. Legal enforcement of fixed money contracts permits each party in a
contract to have sensible expectations that if the other party does not fulfil its
contractual obligation, the injured party is entitled to just compensation and
hence will not suffer a pecuniary loss. Tobin (1985, pp. 108–9) has written
that the existence of money ‘has always been an awkward problem for neoclassical
general equilibrium theory … [and] the alleged neutrality of money
… The application of this neutrality proposition to actual real world monetary
policies is a prime example of the fallacy of misplaced concreteness.’
Tobin then associates Keynes’s rejection of money neutrality presumption
with Keynes’s emphasis on ‘the essential unpredictability, even in a
probabilistic sense’ of the future (Tobin, 1985, pp. 112–13).
The social institution of money and the law of fixed money contracts
enables entrepreneurs and households to form sensible expectations regarding
cash flows (but not necessarily real outcomes) over time and hence cope
with the otherwise unknowable future. Contractual obligations fixed in
nominal terms provide assurance to the contracting parties that despite
uncertainty, they can at least determine future consequences in terms of
cash flows. Entering into fixed purchase and hiring contracts of long duration
limits nominal liabilities to what the entrepreneur believes his or her
liquidity position (often buttressed by credit commitments from a banker)
can survive. Entrepreneurs feeling the animal urge to action in the face of
uncertainty will not make any significant decisions involving real resource
commitments until they are sure of their liquidity position, so that they can
meet their contractual (transaction demand) cash outlays. Fixed forward
money contracts allow entrepreneurs (and households) to find an efficient
sequence for the use of and payment for resources in time-consuming
production and exchange processes.
Money, in an entrepreneur economy, is defined as the ‘means of contractual
settlement’. This implies that in the Post Keynesian monetary theory, the
civil law of contracts determines what is money in any law-abiding society.
In the first page of text in his Treatise on Money, Keynes (1930a, Vol. 1, p. 3)
reminds us that money comes into existence in association with contracts!
The possession of money, or any liquid asset (Davidson, 1982, p. 34), provides
liquidity (a liquid asset is one that is resaleable for money on a
well-organized, orderly spot market). Liquidity is defined as the ability to
meet one’s nominal contractual obligations when they come due. In an uncertain
world where liabilities are specified in terms of money, the holding of
money is a valuable choice (Keynes, 1936, pp. 236–7). Further, the banking
system’s ability to create ‘real bills’ to provide the liquidity to finance increases
in production flows is an essential expansionary element in the
operation of a (non-neutral) money production economy. If tight money
policies prevent some entrepreneurs from obtaining sufficient additional bank
money commitments at reasonable pecuniary cost, when managers (in the
aggregate) wish to expand their production flows (and the liquidity preference
of the public is unchanged), then some entrepreneurs will not be able to
meet their potential additional contractual payroll and materials-purchase
obligations before the additional output is produced and then profitably sold.
Accordingly, without the creation of additional bank money, entrepreneurs
will not be willing to sign additional hiring and material supply contracts and
long-run employment growth is stymied, even when entrepreneurs feel that
future effective demand is sufficient to warrant expansion. A shortage of
money can hold up the expansion of real output, despite expected profits!
Liquid assets also provide a safe haven for not committing one’s monetary
claims on resources when the threat of uncertainty becomes great, as in
Keynes’s discussion of precautionary and speculative motives. Keynes (1936)
claimed that the attribute of liquidity is only associated with durables that
possess ‘essential properties so that they are neither readily produced by
labour in the private sector nor easily substitutable, for liquidity purposes,
with goods produced by labour’.
When agents’ fear of the uncertain future increases their aggregate demand
for ‘waiting’ (even in the long run), agents will divert their earned income
claims from the purchase of the current products of industry to demanding
additional liquidity. Consequently, effective demand for labour in the private
sector declines. Only in an unpredictable (non-ergodic) environment does it
make sense to defer expenditures in this way, as opposed to spending all one’s
earnings on the various products of industry being traded in free markets.
This liquidity argument may appear to be similar to the view of general
equilibrium theorists like Grandmont and Laroque (1976), who stress an
option demand for money. However, in their model and many others, money
has an option value only because of very unrealistic assumptions elsewhere
in the model. For example, Grandmont and Laroque (1976) assume that (i) all
producible goods are non-storable; (ii) no financial system exists, which
means no borrowing and no spot markets for reselling securities; and (iii) fiat
money is the only durable and hence the only possible store of value which
can be carried over to the future. Of course, if durable producible and productive
goods existed (as they do in the real world) and outcomes associated with
holding producible durables were completely orderable, flexible spot and
forward prices would reflect the multiperiod consumption plans of individuals
and no ‘optimizing’ agent would hold fiat money as a store of value. Say’s
Law would be applicable, and the nominal quantity of money would be
neutral. Hence Grandmont and Laroque can achieve ‘temporary’ Keynesian
equilibrium via an option demand for money to hold over time only under the
most inane of circumstances. By contrast, Keynes allowed the demand to
hold money as a long-run store of value to coexist with the existence of
productive durables.
Another approach to liquidity is that of Kreps, whose analysis of ‘waiting’
(Kreps, 1988, p. 142) presumes that at some earlier future date each agent
will receive ‘information about which state prevails’ at a later future pay-off
date. Accordingly, waiting to receive information is only a short-run phenomenon;
long-run waiting behaviour is not optimal in the Kreps analysis –
unless the information is never received! The option to wait is normally
associated with a ‘preference for flexibility’ until sufficient information is
obtained. Although Kreps does not draw this implication, his framework
implies that if agents never receive the needed information and thus remain in
a state of true uncertainty, they will wait forever.
Keynes (1936, p. 210), on the other hand, insisted that decisions not to buy
products – to save – did ‘not necessitate a decision to have dinner or to buy a
pair of boots a week hence or a year hence or to consume any specified thing
at any specified date … It is not a substitution of future consumption demand
for current consumption demand – it is a net diminution of such demand.’ In
other words, neither Kreps’s waiting option nor the Grandmont and Laroque
option demand for money explain Keynes’s argument that there may be no
intertemporal substitution. In the long run, people may still want to stay
liquid and hence a long-run unemployment equilibrium can exist.
This argument has empirical support. Danziger et al. (1982–3, p. 210)
analysed microdata on consumption and incomes of the elderly and have
shown that ‘the elderly do not dissave to finance their consumption at retirement
… they spend less on consumption goods and services (save significantly
more) than the non-elderly at all levels of income. Moreover, the oldest of the
elderly save the most at given levels of income.’ These facts suggest that as
life becomes more truly uncertain with age, the elderly ‘wait’ more without
making a decision to spend their earned claims on resources. This behaviour
is irrational according to the life cycle hypothesis, inconsistent with the
Grandmont–Laroque option demand for waiting, and not compatible with
Kreps’s ‘waiting’ – unless one is willing to admit that even in the long run
‘information about which state will prevail’ may not exist, and these economic
decisions are made under a state of Keynesian uncertainty.
Probabilistic analysis of waiting and option value recognize only a need to
postpone spending over time. However, only Keynesian uncertainty provides
a basis for a long-run demand for liquidity and the possibility of long-run
underemployment equilibrium.
Individuals in the real world must decide whether past experience provides a
useful guide to the future. Should one presume that economic processes are
uniform and consistent so that events are determined, either by ergodic
stochastic processes or at least by specified and completely ordered prospects?
Can the agent completely dismiss any fear of tragedy during the time
between choice and outcome? Does the agent believe he or she is ignorant
regarding the future? No rule can be specified in advance regarding how
individuals decide whether they are in an objective, a subjective, or a true
uncertainty environment. However, their perception will make a difference to
their behaviour.
Keynes laid great stress on the distinction between uncertainty and probability,
especially in relation to decisions involving the accumulation of wealth
and the possession of liquidity. The essence of his General Theory involves
liquidity preferences and animal spirits dominating real expenditure choices.
Money plays a unique role in ‘ruling the roost’ among all assets (Keynes,
1936, p. 223) and it is non-neutral in both the short and long run (Keynes,
1973a, pp. 408–11). These claims, as Keynes made clear in his 1937 restatement
(Keynes, 1973b, pp. 112, 114) of where he saw his general theory ‘most
clearly departing from previous theory’, rested on the clear distinction between
the ‘probability calculus’ and conditions of uncertainty when ‘there is
no scientific basis to form any calculable probability whatever. We simply do
not know.’
Liquidity and animal spirits are the driving forces behind Keynes’s analysis
of long-period underemployment equilibrium, even in a world of flexible
prices. Neither objective nor subjective probabilities suffice to understand the
role of non-neutral money and monetary policy in Keynes’s underemployment
equilibrium analysis. It is not surprising, therefore, that unemployment
still plagues most twenty-first-century economies, since most economists still
formulate policy guidelines which are only applicable to a limited domain
where agents choose ‘as if’ they had specific and completely ordered knowledge
about the future outcomes of their actions.
In Davidson (1978, 1982), I have shown that the existence of the societal
institution of legally enforceable forward contracts denominated in nominal
(not real!) terms creates a monetary environment that is not neutral, even in
the long run. These legal arrangements permit agents to protect themselves to
some extent against the unpredictable consequences of current decisions to
commit real resources towards production and investment activities of long
duration. Legal enforcement of fixed money contracts permits each party in a
contract to have sensible expectations that if the other party does not fulfil its
contractual obligation, the injured party is entitled to just compensation and
hence will not suffer a pecuniary loss. Tobin (1985, pp. 108–9) has written
that the existence of money ‘has always been an awkward problem for neoclassical
general equilibrium theory … [and] the alleged neutrality of money
… The application of this neutrality proposition to actual real world monetary
policies is a prime example of the fallacy of misplaced concreteness.’
Tobin then associates Keynes’s rejection of money neutrality presumption
with Keynes’s emphasis on ‘the essential unpredictability, even in a
probabilistic sense’ of the future (Tobin, 1985, pp. 112–13).
The social institution of money and the law of fixed money contracts
enables entrepreneurs and households to form sensible expectations regarding
cash flows (but not necessarily real outcomes) over time and hence cope
with the otherwise unknowable future. Contractual obligations fixed in
nominal terms provide assurance to the contracting parties that despite
uncertainty, they can at least determine future consequences in terms of
cash flows. Entering into fixed purchase and hiring contracts of long duration
limits nominal liabilities to what the entrepreneur believes his or her
liquidity position (often buttressed by credit commitments from a banker)
can survive. Entrepreneurs feeling the animal urge to action in the face of
uncertainty will not make any significant decisions involving real resource
commitments until they are sure of their liquidity position, so that they can
meet their contractual (transaction demand) cash outlays. Fixed forward
money contracts allow entrepreneurs (and households) to find an efficient
sequence for the use of and payment for resources in time-consuming
production and exchange processes.
Money, in an entrepreneur economy, is defined as the ‘means of contractual
settlement’. This implies that in the Post Keynesian monetary theory, the
civil law of contracts determines what is money in any law-abiding society.
In the first page of text in his Treatise on Money, Keynes (1930a, Vol. 1, p. 3)
reminds us that money comes into existence in association with contracts!
The possession of money, or any liquid asset (Davidson, 1982, p. 34), provides
liquidity (a liquid asset is one that is resaleable for money on a
well-organized, orderly spot market). Liquidity is defined as the ability to
meet one’s nominal contractual obligations when they come due. In an uncertain
world where liabilities are specified in terms of money, the holding of
money is a valuable choice (Keynes, 1936, pp. 236–7). Further, the banking
system’s ability to create ‘real bills’ to provide the liquidity to finance increases
in production flows is an essential expansionary element in the
operation of a (non-neutral) money production economy. If tight money
policies prevent some entrepreneurs from obtaining sufficient additional bank
money commitments at reasonable pecuniary cost, when managers (in the
aggregate) wish to expand their production flows (and the liquidity preference
of the public is unchanged), then some entrepreneurs will not be able to
meet their potential additional contractual payroll and materials-purchase
obligations before the additional output is produced and then profitably sold.
Accordingly, without the creation of additional bank money, entrepreneurs
will not be willing to sign additional hiring and material supply contracts and
long-run employment growth is stymied, even when entrepreneurs feel that
future effective demand is sufficient to warrant expansion. A shortage of
money can hold up the expansion of real output, despite expected profits!
Liquid assets also provide a safe haven for not committing one’s monetary
claims on resources when the threat of uncertainty becomes great, as in
Keynes’s discussion of precautionary and speculative motives. Keynes (1936)
claimed that the attribute of liquidity is only associated with durables that
possess ‘essential properties so that they are neither readily produced by
labour in the private sector nor easily substitutable, for liquidity purposes,
with goods produced by labour’.
When agents’ fear of the uncertain future increases their aggregate demand
for ‘waiting’ (even in the long run), agents will divert their earned income
claims from the purchase of the current products of industry to demanding
additional liquidity. Consequently, effective demand for labour in the private
sector declines. Only in an unpredictable (non-ergodic) environment does it
make sense to defer expenditures in this way, as opposed to spending all one’s
earnings on the various products of industry being traded in free markets.
This liquidity argument may appear to be similar to the view of general
equilibrium theorists like Grandmont and Laroque (1976), who stress an
option demand for money. However, in their model and many others, money
has an option value only because of very unrealistic assumptions elsewhere
in the model. For example, Grandmont and Laroque (1976) assume that (i) all
producible goods are non-storable; (ii) no financial system exists, which
means no borrowing and no spot markets for reselling securities; and (iii) fiat
money is the only durable and hence the only possible store of value which
can be carried over to the future. Of course, if durable producible and productive
goods existed (as they do in the real world) and outcomes associated with
holding producible durables were completely orderable, flexible spot and
forward prices would reflect the multiperiod consumption plans of individuals
and no ‘optimizing’ agent would hold fiat money as a store of value. Say’s
Law would be applicable, and the nominal quantity of money would be
neutral. Hence Grandmont and Laroque can achieve ‘temporary’ Keynesian
equilibrium via an option demand for money to hold over time only under the
most inane of circumstances. By contrast, Keynes allowed the demand to
hold money as a long-run store of value to coexist with the existence of
productive durables.
Another approach to liquidity is that of Kreps, whose analysis of ‘waiting’
(Kreps, 1988, p. 142) presumes that at some earlier future date each agent
will receive ‘information about which state prevails’ at a later future pay-off
date. Accordingly, waiting to receive information is only a short-run phenomenon;
long-run waiting behaviour is not optimal in the Kreps analysis –
unless the information is never received! The option to wait is normally
associated with a ‘preference for flexibility’ until sufficient information is
obtained. Although Kreps does not draw this implication, his framework
implies that if agents never receive the needed information and thus remain in
a state of true uncertainty, they will wait forever.
Keynes (1936, p. 210), on the other hand, insisted that decisions not to buy
products – to save – did ‘not necessitate a decision to have dinner or to buy a
pair of boots a week hence or a year hence or to consume any specified thing
at any specified date … It is not a substitution of future consumption demand
for current consumption demand – it is a net diminution of such demand.’ In
other words, neither Kreps’s waiting option nor the Grandmont and Laroque
option demand for money explain Keynes’s argument that there may be no
intertemporal substitution. In the long run, people may still want to stay
liquid and hence a long-run unemployment equilibrium can exist.
This argument has empirical support. Danziger et al. (1982–3, p. 210)
analysed microdata on consumption and incomes of the elderly and have
shown that ‘the elderly do not dissave to finance their consumption at retirement
… they spend less on consumption goods and services (save significantly
more) than the non-elderly at all levels of income. Moreover, the oldest of the
elderly save the most at given levels of income.’ These facts suggest that as
life becomes more truly uncertain with age, the elderly ‘wait’ more without
making a decision to spend their earned claims on resources. This behaviour
is irrational according to the life cycle hypothesis, inconsistent with the
Grandmont–Laroque option demand for waiting, and not compatible with
Kreps’s ‘waiting’ – unless one is willing to admit that even in the long run
‘information about which state will prevail’ may not exist, and these economic
decisions are made under a state of Keynesian uncertainty.
Probabilistic analysis of waiting and option value recognize only a need to
postpone spending over time. However, only Keynesian uncertainty provides
a basis for a long-run demand for liquidity and the possibility of long-run
underemployment equilibrium.
No comments:
Post a Comment