Sometimes the problem isn’t the data. It’s the way we’re looking at it.
– Tim Noble & Sue Webster – Artwerks, 2020
At first glance it appears to be nothing more than a pile of rubbish: discarded plastic, pieces of metal, random fragments assembled into apparent chaos.
Yet when a spotlight is placed in precisely the right position, the shadow cast onto the wall reveals a clear and recognisable image.
The rubbish has not changed. The shadow has not been edited. No information has been added. The only thing that has changed is the frame through which the object is viewed.
What appeared random suddenly reveals structure. What appeared chaotic becomes ordered.
The artist has not merely created a sculpture. The artist has created a lesson about perception. The patterns we observe often depend as much on the frame we use to interpret reality as they do on reality itself.
Science progresses in much the same way.
We often imagine scientific revolutions occurring because someone discovers a new fact. Sometimes that happens. But many of the greatest advances in human understanding arose because someone found a new way to organise facts that were already known.
Copernicus did not create the planets. He changed the frame through which planetary motion was understood.
Newton did not change the movement of the moon or the fall of an apple. He revealed that both could be understood through the same underlying principles.
Darwin transformed vast collections of biological observations that had previously appeared disconnected into a coherent evolutionary framework.
Einstein showed that assumptions so deeply embedded within physics had become almost invisible. By changing the frame, he revealed a deeper order beneath apparent contradictions.
A recurring pattern runs through the history of science.
Apparent complexity often turns out to be a consequence of perspective.
When we discover the ‘natural’ frame from which to view empirical data, structure emerges from apparent chaos.
The question facing economics is whether it may be time to ask the same question of aparent economic complexity.
Economics and the Rational Frame
For more than a century, economics has largely studied human decision-making through the frame of rational choice.
This has undoubtedly been a productive framework. It has generated powerful insights and elegant models.
But it is worth remembering that rationality is not a fact of nature. It is a frame that we have chosen for analysis.
The distinction matters.
Over time, economics has accumulated a growing collection of behavioural anomalies, forecasting difficulties, competing schools of thought, and observations that sit uneasily alongside the traditional rational-agent picture.
The standard interpretation is that human beings deviate from rationality. This view sees deviations as anomalous, and rationality as the ‘natural’ mid-point around which ‘behaviour’ departs.
But there is another possibility.
What if rationality is simply not the frame that reveals the greatest degree of order in human behaviour?
The Scientific Question
At this point, economics encounters a challenge that is as much philosophical as mathematical.
David Hume famously distinguished between what is and what ought to be.
Science concerns itself with questions of what is.
A scientific theory aims to describe reality as it exists, not as we might wish it to exist. Science is empirical, not normative.
Yet discussions of human decision-making often drift toward a different question:
“What should a rational person do?”
That is a legitimate question. But it is not a scientific question within the domain of macroeconomics. Macroeconomics is concerned with predicting the behaviour of actual economies. As such, its primary task is to identify how people actually make decisions ‘in the wild’, rather than how an idealised agent would make decisions.
Even a perfect theory of what an ideal rational agent ought to do tells us nothing about how actual human beings make decisions. That is an empirical question, which can only be answered by studying the decisions actually made by humans. This is currently the realm of behavioural economics, but its findings are too often treated as departures from rationality, rather than as data worth exploring for its own regularities.
If economics seeks to understand and predict the behaviour of real economies, then its starting point must be reality itself. It must start from how humans do make economic decisions, rather than how humans ought to make economic decisions.
The objective cannot be to discover what rational agents ought to do.
The objective must be to discover what humans actually do, when implementing their evolved, hard-wired algorithms, whether rational or not.
A Possible Inference
Throughout the history of science, apparent complexity has often been a sign that researchers are observing a phenomenon through an unnatural frame. Epicycles multiplied because the geocentric frame obscured the simpler underlying structure of planetary motion. The complexity was not necessarily in the phenomenon itself. It was a product of the chosen representation.
Consider a simple thought experiment:
- Suppose that human brains implement some form of information-processing algorithm.
- Suppose further that this algorithm emerged through evolutionary processes.
If that is true, we might reasonably expect the underlying decision-making rules employed by human beings to exhibit a degree of regularity and simplicity. We can expect a relatively simple relationship between inputs (the information used by the algorism) and outputs (the resulting decision or behaviour).
This ‘simplicity’ might not be obvious. It might only appear simple and elegant when we view it through the ‘natural’ frame of reference. Just as the complexity of epicycles falls away when we view planetary motion from a heliocentric frame. Simplicity becomes visible when viewed through the natural frame of reference.
This leads to an interesting possibility:
If human behaviour appears persistently noisy, unpredictable, and anomaly-ridden when viewed through the lens of rationality, that may not be evidence that human behaviour is fundamentally chaotic. It may instead be evidence that we are looking through the wrong frame.
After all, planetary motion appeared bewilderingly complex under a geocentric frame and surprisingly simple under a heleocentric frame.
The planets did not change.
The lens did.
The Future of Economics
The great challenge for economics may therefore be neither to defend nor reject rationality. It may be to ask a more fundamental question:
- What frame reveals the greatest degree of order in observed human behaviour?
- What frame provides the greatest predictive power?
- What frame allows patterns that currently appear as anomalies to reveal themselves as regularities?
The history of science suggests that these questions are worth asking.
Sometimes the structure is already there. Sometimes the data already contains the answer. Sometimes what is missing is the discovery of the ‘natural’ frame through which order arises.
My own macroeconomic research project has been exploring evidence for one such behavioural regularity that appears remarkably stable across large populations. While the details will have to wait for a forthcoming paper, the experience has reinforced a suspicion: some of the apparent disorder in human economic behaviour may be a consequence of the lens through which we have chosen to examine it.
If so, the future of economics may not lie in discovering new anomalies. Nor building ever-more complex systems to account for the ‘economic epicycles’ we see in behavioural economics.
It may lie in discovering the natural frame that reveals order in human cognition.