I found this a really useful introduction to economics and the various theories and models that have been used in the past, together with new insight drawing on other disciplines.
Traditional Economics: Wealth
is created when people take raw materials from their environment and turn them
into things people want. The more items made in a certain time period, the
richer their creator will be. Division of labour and specialisation of tasks
increases productivity. Best to put resources to the most efficient use; wasting
resources is morally unjust; achievable through a combination of self-interest
and a competitive market. Later theories:
Overpopulation and starvation would adjust food production and prices. Diminishing Returns (a) over-used land
eventually becomes exhausted and (b) at some point you have enough of an item
for your needs.
All theories and models rely on people being logical and consistent in
their behaviours and most ignore the time factor. Supply and Demand is only approximate. Firms keep back stock to increase
demand while service businesses (e.g. lawyers) use staff at less than 100% of
their time in order to manage fluctuations in demand. One Price is an approximation that often breaks down. Different
prices in various countries are not solely down to transportation costs and
trade barriers.
Traditional economics has (a) performed poorly in predicting growth and
recession and in explaining events and (b) misused metaphors and misapplied
theories from other sciences. Economies are complex adaptive systems, and not
closed systems with a predictable end state as the models assume. Computer
modelling and simulations and insights from psychological and scientific
experiments give new insight into economics.
Sugarscape: the computer
island is a 50x50 square grid, with single resource (sugar) in different
amounts per square; two sugar mountains in diagonally opposite corners and a
low or no sugar region between. Each agent (person) only able to look for, move
and eat sugar; at each turn in game can (a) scan for closest unoccupied square
with most sugar within field of vision; (b) move to that square and eat the
sugar; (c) sugar eaten not needed by metabolism = body fat/’savings’; (d) use
more metabolism than eaten, ‘starve’ and die. Agents have pre-set life span, randomly
set differing levels of vision (how many squares they can see), and metabolic
rates. Eaten sugar regrows at one unit per time period. Initially chaotic; agents
in the low/no sugar areas die off and rest quickly move to a sugar rich areas
leaving central plain empty. Each ‘tribe’ efficiently grazes its sugar crop.
Initially fairly egalitarian distribution with large middle class, but then skews
to a few super-rich agents, long-tail of ‘upper-middle class’, shrinking
‘middle class’ and big growing underclass of poor agents. This Pareto curve distribution
is an emergent property of the system: no simple cause-and-effect relationship
driving poverty and inequality.
Tag agents as male or female, and allocate a child-bearing age and a
fertile period; agents in adjacent squares have a baby, with random selection
of abilities from each parent. Baby born in a square next to parents so starts life
in either sugar rich or sugar poor environment. Result: (a) least fit agents
die off, (b) population swings and (c) gap between rich and poor widens
further.
Add new item, spice, in two ‘no sugar’ corners. Agents now need both
items to survive, in varying amounts, but can also trade any surplus; this made
society richer but still wealth variations and geographic clustering in trading
networks. Combination of geography and population dynamics (a) creates heavily
trafficked trading routes, (b) prices never reach equilibrium but fluctuate and
(c) more trading volume than strictly required by logistics, as in the real
world. Time and geographical proximity affected all outcomes, which emerge
bottom up, from the simple starting rules.
The economy is a complex non-linear dynamic system, sensitive to
initial conditions and path dependent (history matters). In dynamic systems,
stock levels and item flows subject to feedback (positive reinforces
connections; negative damps them down) which is affected by time delays,
leading to oscillation round a point. Boom and bust cycles in many commodities,
but their cyclical swings in prices and industry capacity more volatile than
swings in underlying demand or in economy overall; cycles neither quite regular
nor quite random. Responses to demand (increase production and/or raise prices)
have inbuilt time factors.
Economic decisions are based on the information used and types of
decision made. People not always rational or consistent in decision making (take
information available and do the best we can). Our sense of fairness and
reciprocity prompt us to punish people who treat us unfairly and reward people
who help us and give us things – we are conditional co-operators and altruistic
punishers.
People also make mistakes. How a question is framed can affect our
response. Draw big conclusions from small and biased samples. Make decisions on
easily available information instead of finding the important data. Most people
find it difficult to assess probabilities and risks. Tend to look for the most
proximate causes and often confuse random chance and cause and effect. Traditional
economics treats all money the same but people tend to put money into different
mental compartments.
Human mind not brilliant at calculating long equations but we are great
storytellers and story listeners and excel at two aspects of pattern
recognition (a) relating new experiences to old patterns through metaphor and
analogy and (b) very good pattern completers, filling in gaps of missing
information. We develop rules of thumb to move from current state to desired
one, keep track of success, use historically successful rules more than
unsuccessful ones, and learn over time.
A simulated trading environment set up with single stock paying a
random dividend. One hundred agents could buy and sell stock, basing decisions
on (a) historical price pattern, (b) historical dividend pay-out and (c) a
risk-free interest rate. Each agent had 1 rule of thumb, later increased to 100
rules. Single rule simulation results close to traditional economics
equilibrium prediction but 100 rule simulation showed big increases in trading
volume and volatility, with bubbles and crashes, reflecting actual financial
markets much more closely.
Networks essential in any complex adaptive system but glossed over by
traditional economics. [E.g. A thousand
buttons scattered on floor. Randomly join two buttons with thread. At first
lots of two-way connections, then networks, and later at ‘tipping point’, super-networks.]
Two opposing forces in organisations: informational economies of scale from
node growth and diseconomies of scale from build-up of conflicting constraints.
Hierarchies can enable networks to reach larger sizes before diseconomies of
scale set in.
Depressions, recessions and inflation are not exclusively modern
phenomena; irregular historical patterns of little use in predicting economic
behaviour. Complexity economics sees economic patterns as emergent phenomena
that arise out of interactions within the system. Oscillations are a common
feature of complex adaptive systems. Stock markets are much more volatile than
traditional economics predicts; the volatility follows the pattern of a power
law. [There is no typical earthquake
size; they occur across all size scales, but the bigger they are, the rarer
they are.]
Prisoners Dilemma: two suspects in separate cells each told that if
they testify against the other, they will be released, provided the other does
not testify against them. If both testify, each gets reduced sentence. If both
refuse to testify, neither faces jail as evidence not strong enough. While last
option is the best, typically both testify as they cannot communicate with each
other. Various simulations have explored variations on this and the Game of
Life.
Cheaters (mostly) never win and winners (mostly) never cheat. Humans
started with cooperative hunting bands, but big change came with settled
agriculture. The need to divide the resultant wealth led to ‘Big Man Society’ with a political
leader; he would organise work, and specialisations, and take a cut of the
outcomes for his contribution ending up with better housing, food and clothing.
As societies got larger, the Big Man needed others to take on some tasks,
typically allocating them to kin. Trading networks tend to develop first and
most strongly within tribal, ethnic and religious groups. The ugly side of this
is discrimination: classifying some as outsiders. The rule of law enabled
strangers to cooperate, assisted by language as a means of communication.
Market economies took over from Big Men about 300 years ago. Market-oriented
societies are not perfect but their strength is in enabling innovation and
growth.
The absolute level of wealth has an impact on happiness but not in a
linear way. The poor and struggling for survival tend to be less happy, but once
basic needs are met, the correlation between wealth and happiness decreases.
What it means for business and society. All competitive advantage is
temporary, albeit with varying time spans. There is no such thing as a safe,
stable industry. The best you can do is to run faster than the competition. Companies
are Big Men hierarchies and markets are evolutionary machines. In business,
build a portfolio of strategic options. Rigid leaders do best when less
frequent but abrupt changes occur; flexible leaders perform better in volatile
situations.
Market simulation rules: Agent A follows buy low, sell high; Agent B follows
trends and buys rising and sells falling (this amplifies fluctuations); Agent C
is a seasonal trader, and buys and sells on an alternating pattern; Agent D is
a technical trader, with a set strategy to follow. Initially technical traders
did not affect price much, but they soon picked up the oscillating patterns and
made money. As they made bigger trades, they dampened the price fluctuations. Then
volatility suddenly exploded as the large trades introduced their own movements
into the pattern. There are no magic formulas to getting rich.
Politics and policy: complexity approach to economics has the potential
to make the historical framing of politics obsolete. While the Left views
humans as intrinsically altruistic, and the Right that they are intrinsically
self-regarding, they are actually conditional co-operators and altruistic
punishers. Cultures that live for today (or are mired in the past) have
problems (low work ethic, inability to cooperate and low levels of innovation)
while cultures with an ethic of investing for tomorrow value work, have high
intergenerational savings rates, and high levels of cooperation
Income redistribution does not address any behavioural issues (genetic
or cultural) and a laissez faire attitude dooms many to a lifetime of poverty.
We should instead ask ourselves the question ‘If we did not know anything about our draw in the birth-lottery, what
kind of system would we want?’ The answer is a system that combines
equality of upside opportunity with a downside social safety net. It is a
challenge for countries with large immigrant populations to engender trust and
cooperation in a multi-ethnic, multi-cultural society. The ideal is a common
layer of strong norms broadly shared by the society, alongside a further range
of norms, traditions and beliefs.
END
