Humans as information economisers (II)

Satisficing economisers will develop money, suffer money illusion, and not maximise their expectations.

Christine Roy, Unsplash.

This continues my previous post.

Uncertainty ensures herd behaviour

It is helpful in understanding asset markets if we distinguish between positive and negative uncertainty. Negative uncertainty is when insufficient information to calculate probabilities (what is known as Knightian uncertainty) generates fear of loss. Positive uncertainty is when such uncertainty generates hope of gain.

The direction of uncertainty depends on what people read into what cannot be calculated. In such a situation, the most salient information turns out to be what other people are doing. Especially given the value of assets (as things held across time) depends so much on expectations about their future value.

So-called tech booms, which have been a feature of technologically innovative societies all the way back to the 1840s railway boom, are examples of positive uncertainty. Such booms often end when more information becomes available about the commercial implications of the new technology. The drop in uncertainty due to increased information leads to less overall investment in that class of assets, rather than more.

Uncertainty is an irreducible element in asset markets. In part because probability calculations often rest on assumptions about the distribution of possibilities.

Where there is not enough information to calculate probabilities, then there is also no solid information for independent judgement. What John Maynard Keynes called “animal spirits” is imitative behaviour in the absence of calculable probabilities.

By acting as others are, one can “piggyback” off other’s (presumed) information while minimising the risk of doing significantly worse than others. In other words, use the same strategy that generates flocking and herding behaviour in birds and animals.

That uncertainty cannot be eliminated means that there is always some herd behaviour in markets. Behaviour that is not, in itself, irrational nor implies that markets are inferior mechanisms for dealing with, and generating, information.

Adapting the model of Barberis et al, and taking an asset-price bubble as being when the price of an income-producing asset-class has surged beyond its expected income value (“market fundamentals”), then such asset-prices have moved out of the realm of calculable risk and into the realm of uncertainty. In that case, among those willing to trade in such an asset, the strongest signal is going to be what others are doing. So, expectations will be extrapolated from the current path of the asset value.

Any variation in “wavering” between that signal and expected income is going to produce room for trades. Once a turning point is reached, the price can be expected to collapse back to expected-income value. Positively-read uncertainty becomes negatively-read uncertainty, with the value collapsing back to (or perhaps below) calculable risk. In such circumstances, wealth simply vanishes.

If there is no surge in asset-prices beyond expected income value, then we do not have a bubble, just asset-price volatility. Even if there is a bubble as defined above, that such turning points are not predictable (requiring information not yet available, so not yet feeding into expectations, so prices) means that the identification of such a bubble is not very useful. Moreover, the history of monetary authorities targeting asset-values (in either direction) is not a happy one.

Housing is a somewhat distinctive case, as rents (so asset income-values) tend to adjust to scarcity and land value (frequently driven by regulatory restrictions) is often most of the value of housing. For more on the interaction between rents and house prices, read Kevin Erdmann.

About assets

Assets have wealth and output/income/resilience value. The latter is sometimes expressed as fundamental value, as if marginal analysis had never been discovered. Fundamental, like real, is a metaphysically portentous term economists should avoid using. (What is a “real” economy without money? Much smaller as there is way more transaction friction, so fewer transactions.)

The output-value of an asset is its value in producing goods and services. The income-value of an asset is the value of its stream of income. The resilience-value of an asset is its ability to adjust to changes in circumstances: i.e., how robust the benefit(s) that asset provides are to changing circumstances.

Resilience will be more valued the higher the level of risk, and especially negative uncertainty, is. For instance, gold is scarce, dense and very physically stable across time, so has been a perennial resilience asset. The interactive combination of output- income- and resilience-value produces the core-value of an asset.

The wealth value of an asset is what people will pay for it as a store of value. This can fluctuate wildly, for it is highly dependent on interactive expectations and so on shifts in information. The common expectation that an asset will rise in value makes it more attractive as a store of value, driving its wealth-value up. Conversely for expectations of a fall in value.

As asset prices are so affected by expectations of future value, and it is entirely possible for wealth to just vanish, asset markets are hungry for information.

The conventional two-axes, price and quantity, graphs for supply and demand for goods and services are somewhat inadequate for assets, as they are held across time and their price is so dependent on expectations about their future value.

Moreover, as assets are held across time, they have management costs. Biological organisms have various thresholds they generally try to avoid crossing, such as death. Though lineage persistence can trump that: for instance, in a male praying mantis making a gift of its own body for consumption by the female for the opportunity to reproduce.

Risk-tolerance can be reasonably thought of how averse to variability in outcomes one is. Peasants in environments where soil quality and rain are both variable, and the price of time (aka interest rates) are high, will seek to operate dispersed fields, even though that reduces their average production, so as to avoid falling below the threshold of successful subsistence. That is, they will sacrifice efficiency (average level of production) for resilience (avoiding falling below the threshold of subsistence).

Management of assets also imposes costs in terms of acquiring and processing information. The higher the information demands in managing an asset, and the higher the variability in value and return an asset is, the higher the management cost. The higher the management cost, the higher the (gross) return of an asset has to be to be worth holding.

Thus, the equity premium puzzle is only a puzzle if you do not fully incorporate time into your analysis. It is a milder version of the phenomena that generates enormous profits from illegal narcotics, due high level of risks and information management demands. Such risks include transactions and assets that have to be defended (and are more likely to be contested) by private action, due to their illegality.

Reducing transaction friction

Money existed before states existed. Societies have used many different things as money. The following is a list of things used as money compiled from A. H. Quiggin’s A Survey of Primitive Money:

Beads (whether made of stone, shell, glass or whatever), beeswax, buffaloes, camphor, cattle, cloth (from silk to wadmal, a coarse wool fabric), coca leaves, cocoa beans, coconuts, strings of coconut discs, dye cakes, feather coils, gold dust, weighted gold, grain (notably barley and rice), human heads, logwood (mahogany), tool metals (iron, copper, tin, bronze) in various shapes, plaited palm-fibre rings, pigs, porcelain jars, balls of rubber, salt (including stamped salt cakes), seeds, shells (especially cowries), silver in lumps or shaped, animal skins, slaves, carved stone (a tool material), tea (including in bricks), teeth, tobacco.

These things have all been used a money because they have been used directly in exchange for goods and services or they have been used to discharge, equalise or otherwise quantify obligations.

Something is being used as money when you are using it for its value in exchange, not because of some other feature it might have. To put it more formally: something is being used as money when its use-value is its exchange-value.

That we are highly imitative, monkey-see, monkey-do, species aids the rise and use of money. That we are a highly normative species means that we have property (acknowledged possession) and exchange (acknowledged switch in possession).

The great advantage money provides is reducing friction in transactions by economising on search. You don’t have to find someone who has what you want and wants to swap what you have for it. Instead, you just offer them money.

The point of money is to reduce transaction friction (the rate at which transaction costs slow down transacting). Economising on cognitive calculation helps transaction friction, so there will be a trade-off between cognitive calculation and ease of transaction.

The level of liquidity a good or asset provides is how low its transaction friction is.

Indeed, the reducing-transaction-friction value of money (what we might call its transaction utility) is so great, folk continue to use money even during hyperinflation, when money is, by orders of magnitude, the worst store of value in the economy.

Money minimises transaction friction, including cognitive effort. There is a powerful incentive to do or use what reduces transaction friction, as that economises on time and effort. Hence cultures persistently create money, they persistently come to settle on one or more things to commonly use as an exchange good. That is, a good whose exchange-value becomes its use-value (i.e. money).

Financing will tend to be done using that which minimises transaction friction. This is so, not merely to economise on time and effort, but because such minimising increases responsiveness and flexibility. This is important in dealing with risk, coordinating and managing resources and responding to discovery: the central activities of commerce.

Money operates across time, between production of x and consumption of y. Which, as money can be held across time, can be in quite different time periods. If your model does not incorporate time, it cannot incorporate money (nor, of course, discovery).

Money illusion comes from us economising on information to the extent of over-estimating the stability of the exchange-value of money. Whether because we are unused to the value of money shifting significantly or because the benefit from the required calculations is not worth the extra effort.

Using the term real to talk about output, abstracting away from money, is to imply that money is not a good. Money is a good: one that provides exchange services that reduce transaction friction.

Most trade is carried by water because water generates a great deal less friction than land. Most exchange is monetised because money generates a great deal less transaction friction than barter (in-kind exchange). Much muddy thinking about money comes from thinking of it as being like water that carries trade (“medium of exchange”) rather than as an exchange good that can also be an asset.

What flows between people is resources (goods and services), some by coercion (such as taxes and transfers), some by exchange, some via investments in connection (notably gifts), some via sharing (which can be trading across time). Flows that often incorporate money as an intermediary good, providing exchange services, but by no means always do.[#]

Inflation and deflation are still monetary phenomena. Inflation is where the use of money in exchange is rising relative to output. Deflation is where the use of money in exchange is falling relative to output. In the former case, the exchange-value* of money is falling. In the latter, it is rising.

As money can be used as an exchange good or held for future use (i.e. be an asset), it is the hinge between goods and services and assets. It can thus be used as a unit of quantification across goods, services, assets (including debts) and other obligations (i.e., be a unit of account).

In subsistence economies, people worry about having enough stuff (food, clothing, shelter). In monetised economies, people worry about having enough money—the good whose exchanges services enables access to all the goods, services and assets available if one pays.[&]

Money is the photon of commerce: both an exchange good reducing transaction friction and an asset held for future use. To understand the effect of money, it is important not to treat those two uses as a single quantity while recognising that shifts between those uses matter.

If the exchange value of money is falling, the incentive is to spend it rather than hold it. The more money is used in exchange relative to output, the more its exchange value will tend to fall.

This is why inflationary expectations matter. In the case of hyperinflation, inflationary expectations become so entrenched, the hyperinflation only ends when the previous money stops being used. Either at all (as in Confederate dollars at the end of the US Civil War) or by being replaced by a new currency.

Deflationary expectations also matter. If folk expect the exchange value of money to rise, the incentive is to hold it as an asset for future (more valuable) use. The less people use money in exchange, the more its exchange value is expected to rise, the more the incentive to hold it for future use, the less spending.

The less spending, the lower incomes, the higher the burden of debts, so the more people default, so the more financial institutions find their loan assets vanishing, so the more they collapse as their liabilities surge past their assets. This is the debt-deflation spiral.

The Bank of France continually took gold out of monetary use (i.e., hoarded it) from 1928 onwards by undervaluing the franc in terms of gold. Gold therefore bought more in France than elsewhere. Especially than in the UK, whose pound was overvalued, so gold bought less there than elsewhere. Gold therefore tended to flow out of Britain and into France, an interaction that adversely affected Anglo-French relations.

The Bank of France’s gold holdings went from seven per cent of total world gold reserves in 1927 to 27 per cent in 1932 without it making any other adjustments to monetary policy, so such gold just exited the system. This gold hoarding, in the absence of countervailing action by the US Federal Reserve, meant that the supply of gold operative in the monetary system shrank, so the price of gold continually rose.

Under the gold standard, gold set the exchange-value of money, which thereby kept rising, so prices kept falling, so spending kept falling, so incomes kept falling, so … . This created the debt-deflationary spiral that was the Great Depression. The Bank of France (with some help from the US Federal Reserve) had turned the gold standard into a deflationary-death-spiral. Economies exited the Great Depression as they exited the gold standard.

Once the US Federal Reserve had been created in 1913, so that four central banks (the Fed, the Bank of England, the Reichsbank and the Bank of France) dominated gold holdings, the gold standard turned into a fully technocratically managed system, rather than the self-calibrating system gold standard enthusiasts envisage. Yes, the Great Depression was a case of technocratic incompetence with global effect (as was the later Great Recession of 2007-9).

The proportion of money held across time periods (i.e. as an asset) is the reciprocal of how quickly it circulates in transactions. The more it is held, the slower its rate of circulation (what economists call “velocity”). The less it is held, the faster its rate of circulation.

In hyperinflation, no one holds money more than they absolutely have to, so it circulates faster and faster. Hence its rate of circulation heads ever higher as its exchange-value heads ever lower.

Any claim that the level of transactions, so of production, distribution and exchange, is indifferent to the level of transaction friction in the economy is obviously false. The use of money matters, as does the regulatory structure and whether one has rule of law. One of the perennial problems of Latin America, for instance, is regulatory structures that greatly increase transaction friction, so reduce the level of transactions, so of economic activity.

For money to be neutral over some time period requires there to be no change in the level of transaction friction and there be no other effect on output from any changes in the use of money in an economy. As money is an exchange good, providing exchange services, with a price relative to output (aka the price level, P), this is clearly not true in the short run, as the effects of inflation (continual rises in P) and of deflation (continual falls in P) demonstrate. Given the level and quality of debt matters, as does the price of time (aka interest rates), affecting action across time periods, and both debt and interest rates are affected by use of money as an exchange good and as an asset, money is not going to be neutral in the long run either.

From the uses of money as an exchange good and as an asset flow money being a unit of account, a medium of exchange, a medium of account (i.e., both) and a store of value. All these features are consequences of money being something whose use-value is its exchange-value and money being able to be held as an asset, they are not fundamental features of money.

Any asset is a store of value, for instance, and the role of money as an asset is, as is normal with assets, not just being a store of value. Treating these features as fundamental, as the starting point for analysing money, muddies thinking about money.

As money is something where its use-value is its exchange-value, it can be cash (coins, notes), it can be some commodity (see above list), it can be notes exchangeable for some commodity (as in the gold, silver or bimetallism** standards) or it can transferrable account entries.

A crypto-currency such as Bitcoin is money (something whose use value is its exchange value) created by an algorithm in a distributed network. As the supply of the crypto-currency is the distilled record of its past use, it is not much responsive to current demand (i.e. its supply is highly inelastic). Hence the exchange value of crypto-currencies tends to be highly volatile.[&]

Leaving aside crypto-currencies, transferrable account entries are bank-created money.[#] Economists have entertained three theories of such money. In the words of economist Richard Werner:

According to the financial intermediation theory of banking, banks are merely intermediaries like other non-bank financial institutions, collecting deposits that are then lent out. According to the fractional reserve theory of banking, individual banks are mere financial intermediaries that cannot create money, but collectively they end up creating money through systemic interaction. A third theory maintains that each individual bank has the power to create money ‘out of nothing’ and does so when it extends credit (the credit creation theory of banking). The question which of the theories is correct has far-reaching implications for research and policy. Surprisingly, despite the longstanding controversy, until now no empirical study has tested the theories. (Last emphasis added.)

That banks simply create money with every loan and have to balance assets (their loans) with their liabilities (people’s deposits), has become, once again, the most widely accepted theory of bank money.

Economists spent a century or so ending up where they started in “explaining” what banks do, without, for decades, doing the field work. This is not science, this is intellectual parlour games.

Footnotes

* Relative to output. Currencies also have exchange value with other currencies: nowhere in this essay does that usage apply.

** Bimetallism is where a currency has a set value in gold and one in silver. One can reasonably argue that the 1823-1873 system, where some countries were on the gold standard, some on the silver standard, some on bimetallism, was more stable than almost all major economies being on the gold standard.

[#] Edited since original posting to clarify the points being made.

[&] Added paragraph.

References

Books

Peter L. Bernstein, Against the Gods: The Remarkable Story of Risk, John Wiley & Sons, [1996] 1998.

Charles W. Calomiris and Stephen H. Haber, Fragile by Design: The Political Origins of Banking Crises and Scarce Credit, Princeton University Press, 2014.

Edward Chancellor, The Price of Time: The Real Story of Interest, Penguin, 2022.

R. H. Coase, The Firm, the Market and the Law, University of Chicago Press, [1988] 1990.

George Cooper, The Origin of Financial Crises: Central Bank, Credit Bubbles and the Efficient Market Fallacy, Harriman House Ltd, [2008] 2010.

Michael J. R. Crawford, An Expressive Theory of Possession, Hart Publishing, 2020.

Irving Fisher, The Money Illusion, Start Publishing, [1927] 2012.

James Franklin, The Science of Conjecture: Evidence and Probability before Pascal, Johns Hopkins University Press, [2001] 2015.

Frank H. Knight, Risk, Uncertainty and Profit, Cosimo, [1921] 2005.

Carmen M. Reinhart & Kenneth S. Rogoff, This Time Is Different: Eight Centuries of Financial Folly, Princeton University Press, 2009.

Scott Sumner, The Money Illusion: Market Monetarism, the Great Recession, and the Future of Monetary Policy, University of Chicago Press, 2021.

Scott Sumner, The Midas Paradox: Financial Markets, Government Policy Shocks, and the Great Depression, Independent Institute, 2015.

Benoit Mandelbrot with Richard L. Hudson, The (Mis)behaviour of Markets: A Fractal View of Risk, Ruin and Reward, Profile, [2004] 2008.

A. Hingston Quiggin, A Survey of Primitive Money: The Beginnings of Currency, Routledge [1949], 2019.

Carlo Rovelli, The Order of Time, (trans.) Erica Segre and Simon Carnell, Penguin, [2018] 2019.

George Selgin, Good Money: Birmingham Button Makers, the Royal Mint, and the Beginnings of Modern Coinage, 1775–1821, Independent Institute, [2008] 2011.

Nassim Nicholas Taleb, Antifragile: Things That Gain From Disorder, Penguin, 2012.

Articles, papers, book chapters, podcasts

George A. Akerlof, ‘Labor Contracts as Partial Gift Exchange,’ The Quarterly Journal of Economics, Vol.97, No.4 (Nov. 1982), 543-569.

Nicholas Barberis, Robin Greenwood, Lawrence Jin, Andrei Shleifer, ‘Extrapolation and bubbles,’ Journal of Financial Economics, 2021, 129, 203–227.

Truman F. Bewley, ‘Why Not Cut Pay?,’ Cowles Foundation Discussion Papers, 1997, 1415.

Claudio Borio and Piti Disyatat, ‘Global imbalances and the financial crisis: Link or no link?,’ BIS Working Papers, No 346, May 2011.

Claudio Borio, Piti Disyatat, Mikael Juselius and Phurichai Rungcharoenkitkul, ‘Why so low for so long? A long-term view of real interest rates,’ BIS Working Papers, No 685, December 2017.

Bruce Champ, ‘Private Money in our Past, Present, and Future,’ Economic Commentary, Federal Reserve Bank of Cleveland, January 1, 2007.

Satyajit Chatterjee, ‘A Theory of Asset Price Booms and Busts and the Uncertain Return to Innovation,’ Philadelphia Fed Business Review, Q4 2011, 1-8.

Alan Page Fiske, ‘The Four Elementary Forms of Sociality: Framework for a Unified Theory of Social Relations,’ Psychological Review, 1992, Vol. 99, No. 4, 689-723.

Chris D. Frith, ‘The role of metacognition in human social interactions,’ Philosophical Transactions of the Royal Society B, 2012, 367, 2213–2223.

Herbert Gintis, Carel van Schaik, and Christopher Boehm, ‘Zoon Politikon: The Evolutionary Origins of Human Political Systems,’ Current Anthropology, Volume 56, Number 3, June 2015, 327-353.

Milton Friedman, ‘The Role of Monetary Policy,’ American Economic Review, Vol. 58, No. 1 (Mar. 1968), 1–17.

Milton Friedman, ‘The “Plucking Model” of Business Fluctuations Revisited,’ Economic Inquiry, Western Economic Association International, vol. 31(2), April 1993, 171-177.

Chris D. Frith, ‘The role of metacognition in human social interactions,’ Philosophical Transactions of the Royal Society B, 2012, 367, 2213–2223.

Cameron Harwick, ‘Money and its Institutional Substitutes: The role of Exchange Institutions in Human Cooperation,’ Journal of Institutional Economics, 2018, 14(4), 689-714.

Hillard Kaplan, Jane Lancaster & Arthur Robson, ‘Embodied Capital and the Evolutionary Economics of the Human Life Span’, in Carey, James R. and Shripad Tuljapurkar (eds.), Life Span: Evolutionary, Ecological, and Demographic Perspectives, Supplement to Population and Development Review, vol. 29, 2003. New York: Population Council, 152-182.

Hillard Kaplan, Kim Hill, A. Magdalena Hurtado, and Jane Lancaster, ‘The embodied capital theory of human evolution,’ in Peter T. Ellison (ed.) Reproductive ecology and human evolution, De Gruyter, (2001), 293-317. Aileen Oeberst and Roland Imhoff, ‘Toward Parsimony in Bias Research: A Proposed Common Framework of Belief-Consistent Information Processing for a Set of Biases,’ Perspectives on Psychological Science, 2023, 1–24.

Arnold Kling, ‘Macroeconometrics: The Science of Hubris,’ Critical Review, 2011, 23:1-2, 123-133.

Douglas A. Irwin, ‘The French Gold Sink and the Great Deflation of 1929–32,’ Cato Papers in Public Policy, Vol.2, 2012.

Michael McLeay, Amar Radia and Ryland Thomas, ‘Money in the modern economy: an introduction’, Bank of England, Quarterly Bulletin, 2014 Q1, 4-13.

Michael McLeay, Amar Radia and Ryland Thomas, ‘Money creation in the modern economy’, Bank of England, Quarterly Bulletin, 2014 Q1, 14-27.

James Tobin, ‘Commercial Banks as Creators of “Money”,’ in Banking and Monetary Studies, Dean Carson (ed.), US Treasury, 1963, 408-419.

Richard A. Werner, ‘Can banks individually create money out of nothing? — The theories and the empirical evidence,’ International Review of Financial Analysis 36 (2014) 1–19.