Tuesday, March 29, 2016

Structuring the unstructured: A pluralist economics mud map

In a previous article (“Reforming Economics: The Challenge”), I made the point that organising the jumbled schools of economic thought into a coherent pluralist curriculum faces both a social and a technical challenge. These two challenges go hand in hand to some degree, since the teaching within any discipline largely reflects the sociology of its practitioners. Social conventions are reflected in teaching, and teaching reinforces those social conventions.

In economics, and the undergraduate courses where the majority of students get their complete economics training, this means uniform domination by the neoclassical approach – mirroring its dominance in mainstream journals and research activities. Even the now decades-old behavioural and experimental school is all but ignored in core economic textbooks, reflecting a contested relationship with the mainstream and an inability to escape the fringes of the discipline.

But I hold an optimistic view. There is a feedback loop that can be broken by offering an attractive teaching alternative that presents an array of approaches and allows for conflicting ideas and methodology to be examined side by side. Such an alternative needs a very broad conceptual map on which each school of thought can be placed, and it needs tools to comprehensively understand and assess the validity of each. Such a map can allow a logical positioning of competing methods across domains of economic interest and can illuminate where and when there might be overlaps or conflicts. Such a map could facilitate a broader language of economics by enabling ideas to be accurately communicated between different schools of thought.

Absent a map, teaching a pluralist curriculum faces huge consistency problems. Others have attempted the relatively straightforward task of incorporating behavioural approaches into core economics courses. They find that conflict between ideas is a teaching obstacle and that there is a tendency to pick winners and losers in such circumstances.
…integrating such insights into microeconomics is not easy. One has to tread a fine line between integrating new and important results without making the standard theory look completely useless to students. This is often advanced as an argument not to integrate newer results at all. I tend to disagree with this view. It supposes that students are unable to hold two opposing thoughts in their head, which I find more than a little patronizing.
Readers may agree that standard microeconomic theory is completely useless, but teaching a pluralist approach does require giving credit to the mainstream where it may be relevant. Finding that relevance needs a map.

To illustrate, I’ve drawn a mud-map of the economic terrain in a very broad sense. The boxes on the map below represent islands of interest that are related to each other, but are conceptually different domains that cannot be conflated. Money for example, is not a real resource, or a traded widget, as it is almost universally assumed to be in neoclassical models. It is a set of accounts. It has its own domain. Real resources in the economy are also not the set of legal rights over who owns what. These rights are entirely separate. To be clear, money can be thought of as a type of ownership right, so the money and legal rights domains share a common red ‘ownership’, or `power’, environment in my map.

Welfare is another domain: distinct but related to real resources, the system of legal rights and the monetary environment. It is this domain that provides a moral foundation for the economy. This domain shares its yellow terrain with the real resources, since can both be considered the area in which mainstream economic theory operates, currently being bridged by the rather inadequate welfare theorems.

Outside these economic islands is the sea of the social and political environment. Any economic domain is enclosed within a social environment, and any analysis of economic problems depends crucially on conditions within this environment. Think of this domain as the evolving norms and institutional structures of society.

Floating in this sea are three tools, or ways to interrogate ideas, that can be carried to each economic island to explore and assess different approaches and to guide learning and discussion.

The first tool is timing. We know the world is dynamic, so asking the question of how timing is dealt with in any economic approach is crucial. History is irrelevant in most mainstream models, while the future is certain (though sometimes risky). Equilibrium arises instantly without adjustment. Asking the question of timing provides clarity about what conditions would need to be met in order for an approach to be valid, and why it might break down. It would expose that a number of diverse theoretical approaches have a common ignorance of time, while others (e.g., evolutionary economics) hold it as their core insight.

The next tool is aggregation. At what level of aggregation is analysis taking place? Is it important that individual and aggregate behaviour may differ? This tool is used to avoid the macro-micro distinction and concentrate on relevant economic questions. After all, a single market is already a macro-economy of buyers and sellers. Firms too are aggregates of individuals, and their existence is worthy of discussion in the legal rights domain. Questions about the how and why of aggregation are some of the least-asked but most important in any domain. How can we aggregate welfare or capital? How are rights divided between and within entities? What level of analysis matters in the monetary system?

Lastly, there is the tool of prediction. This represents the ‘So what?’ question that needs to be asked of any method of analysis on every economic island. Even if an analytical approach seems to adequately address questions of timing and aggregation, it can’t pass scientific muster unless it provides useful predictions. If different schools of thought generate different predictions, it should be possible to trace backwards through the questions of timing and aggregation and (being clear about which domains are being examined and how they are linked) to find the causes of those different predictions.

It might help here to provide an example of how this map can aid the teaching of different ideas in economics and facilitate communication between schools of thought. The map could be used in two ways; either as a way to structure courses, by exploring domains and tools and incorporating different approaches where relevant, or as a way of structuring inquiry into different schools of thought as they arise in the curriculum.

If one were using this map for structured learning about Godley and Lavoie’s monetary circuit models, students would first learn about the institutional setup that produces the monetary domain, including what exactly money is, and how certain relationships to the legal rights and real resources domains are implied. They would then note the way timing is treated and how and why firms, government, banks and other sectors of the economy are aggregated. Finally, they would explore the types of predictions and how they are linked between the money, legal rights, resources and welfare domains.

Later, when learning about mainstream growth theory, it would be clear that this approach resides in the real resources domain where time is condensed to a single point, while its representative agent (or ‘social planner’) deals with aggregation. Predictions of such growth models vary, but most generate only static equilibrium outcomes unless some external shock hits the system. Both the monetary and legal domains are assumed to allow for the outcomes of the models. In practice, the capital terminology used in these models is often confused for capital in the legal rights domain.

We could dig down further into the neoclassical view to explore the still controversial question of what is capital. Here we would see that in standard growth models capital is a collection of physical objects. Yet because we cannot aggregate quantitative measures of the diverse physical objects, but natural and produces, we end up accounting for them by monetary measure of value that are instead comparative measures of value of property rights, which are constructed by our institutions.

By situating these two approaches on the map, and seeing the differences as to aggregation, timing and predictions, we can more clearly see under what conditions they correspond or conflict. For example, investing in more physical capital generates higher growth in both, but only in a monetary circuit model can we potentially say anything about the adjustment processes in the economy. Further, both take the social and political environment as given and only indirectly relate to the welfare domain through an implicit assumptions about the relationship between aggregate resources and aggregate welfare.

It may take a little work, but this type of structured thinking is helpful.

I am not alone in trying to put some structure around the jumble of economic schools. Ha-Joon Chang has mapped schools of economic thought based on particular domains of interest. In his latest book Economics: The User’s Guide there are many of the same general themes of timing (Economies change through…, The world is…) and aggregation (The economy is made up of…), and also the suggestion that different economic approaches focus on certain areas of the economy, such as production or exchange. 

If we want a viable pluralist economics curriculum, a way to structure ideas from the diverse schools of thought is absolutely crucial, and I hope this mud-map of economic domains can provide a starting point. New pluralist textbooks and teaching materials based on a structured inquiry can demonstrate how diverse ideas can be brought together, without creating conflict and without transforming the exercise into merely a process of selecting a school of thought that aligns with the student’s existing ideologies. I hope this outcome can emerge from the combined efforts of IDEA Economics, Evonomics, Rethinking Economics and other reformers.

First published at IDEA Economics

Tuesday, March 22, 2016

Reforming economics: The Challenge

And so the debate rages. Economics needs to change. Always does.

The challenge of reforming economics cannot by overstated. Modern mainstream economics has remained dominant in our universities and governments despite overwhelming evidence against most of its core principles, and despite decades of attempted revolutions. The concept of a static equilibrium and the ‘representative agent’ method of aggregation are just two notions that have been repeatedly shown to be internally inconsistent; not just by outsiders, but by many of the leaders in the mainstream. Yet they continue to dominate the discipline.

The core remains unchanged.

Outdated and economically-irrelevant concepts still fill the pages of introductory textbooks. From there they fill the minds of each new generation of students, who pass on these ideas to the next generation of students, and across society more broadly. Breaking the feedback loops in this system is what is required to transform the discipline.

The call for pluralism is an admirable end goal, upon which most reformers agree. But my view is that previous attempts at reforming economics have failed because they avoided, or inadequately understood, the two main barriers to change. While they may at first seem insurmountable, without leveraging change at these points the mainstream will stay locked-in as the dominant approach to economic analysis.

The first main barrier is social. Economics as a discipline typically rewards tribalism over reconciliation. If you’ve been following economics blogs in recent years you will have a pretty clear understanding of this. But the same dynamic happens in all of academia. Journals are often aligned with particular views on what are acceptable methods and concepts and act as the gatekeepers to the tribe, requiring all comers to offer sacrifices to tribal elders. Moreover, the mainstream represents over 80% of the discipline, so any change promoted by minority groups will be frowned on. This is the social reality.

In essence, the social challenge is to bring the tribes together in a way that makes them all feel like insiders in a new larger group. This means not starting fights with powerful tribes, especially not the current mainstream. It means highlighting any common ground where tribes agree and giving credit to how they contribute to an enhanced view of economics.

I will talk more about social barriers to change in economics at length in the future. For now though, this is enough context to discuss the second main barrier, which is a technical one.

The technical problem is - how do you teach a pluralist program when there is no recognised structure for presenting content from many schools of thought, which can often be contradictory, and when very few academics are themselves sufficiently trained to to so?

Teaching a pluralist curriculum shouldn’t be about presenting the economics discipline as one of feuding tribes. I share Simon Wren-Lewis's fear that a pluralist curriculum could become a one-stop shop for students, who get to browse the tribes before joining the one that most aligns with their existing political ideology.

Instead, I sincerely hope that we can train a generation of economists to be aware of the legacy of each school of thought, but acknowledge the common ground between them.

There is an old saying that if you ask five economists and you'll get five different answers - six if one went to Harvard. Can we teach a pluralist curriculum which would bring economists onto the same page so that when you ask five economists a question, you get one good answer?

Approaching this problem needs a systematic solution. For example, we need to think about how to structure teaching around topics and concepts that allow students to study problems and evaluate potential approaches, and their evidence. We need an alternative textbook, or set of them, that can satisfyingly demonstrate the approach being called for, and ultimately offer a replacement foundation upon which to build a pluralist curriculum.

Presently, even the best mainstream textbooks merely tack on a few comments on alternative approaches. For example, the currently popular experimental or behavioural economics schools, despite being widely being regarded as a revolution in economic thinking and economic science, are given very little credence in the most popular textbooks. After pouring through the text of 25 popular undergraduate microeconomic textbooks, Lombardini-Riipinien and Autio find that
… ten of the 25 textbooks examined make no reference at all to behavioral economics; six dedicate less than 1% of total pages to it, six between 1% and 2.6%, and three between 6% and 11%. When behavioral economics is discussed, the focus tends to be on bounded rationality rather than on bounded self-interest or bounded willpower.

Experimental economics is not discussed at all in ten textbooks, twelve textbooks dedicate less than 0.6% of total pages to it, while three dedicate between 2% and 10% of total pages.
Joan Robinson tried to comprehensively rewrite the core introductory economics textbook with John Eatwell in 1973. While the book does a superb job of putting economic analysis in a philosophical and historical context, it offers no coherent backbone upon which to build an understanding of economics. For example, after reading the book I learnt very little to aid me in answering practical day-to-day questions about the economy. Where does money come from? How do we measure unemployment? How could we assess alternative options for addressing negative externalities? Is the very concept of eternality useful, since it implies the existence of a no-externality world?

What is needed is a way to structure the exploration of economic analysis by arranging around economic problems around some core domains. Approaches from various schools of thought can be brought into the analysis where appropriate, with the common ground and links between them highlighted.

Unless the community of economic reformers can make the effort to reconstruct the way economic is taught, and make the tough decisions about how to structure new core texts, including what to leave in and what to leave out, then change will remain elusive. We can’t call for change in the challenging tribal social environment of economics without offering an attractive alternative - one that embraces the best from each of the schools of thought and finds common ground without creating a new set of outsiders.

Originally published at IDEA Economics

Thursday, March 17, 2016

Queensland housing supply

The planning process is often held to be partly to blame for housing prices in Australia. Unfortunately there is very little data about the planning system in general, and information on the stock of approvals and the number of new approvals is typically even worse.

In Queensland the data has become better over the past few years as Council's have begun reporting to the State their approvals activity, and their estimates for the capacity of their planning scheme to provide new dwellings. Councils report that they currently have land zoned for around 660,000 new detached houses. That's about 40 years worth of new detached dwellings able to be built without any changes to zoning laws or any approvals outside the code. To put that in perspective, the total population growth in Queensland is about 70,000 people per year, who require one home per 2.6 people, or 26,000 new homes per year of both apartments (about 10,000) and detached houses (about 16,000).

There are also current approvals for 107,000 new homes, or around 4 year's worth, and 24,000 new approvals being granted each year.

For more fine grain analysis of the housing supply pipeline, I have created the visualisation below using data from the Queensland Government Statisticians Office. It allows you to see the historical trends in planning approvals for residential dwellings, as well as the stock of current approvals, the lapses of approvals, and so forth. Click on the map to get the historical time series of these indicators for that region. 

Nowhere can I see that councils have hit an approvals limit that might have constrained the rate of supply of new dwellings in an area.