Literary Insights

BOOK LINK:

Here is a summary of the key points from “The World in the Model”:

Over the last two centuries, economics has changed from a social science based on words to one based on mathematical models. This book analyzes this change historically and philosophically through case studies of specific economic models.
It explores how economists create models and reason with them. The book is intended for economists as well as scholars in other fields interested in how economics works from the inside.
Part I discusses how modelling became the dominant method of enquiry in economics. It naturalized the use of models, moving from verbal “laws” to formal representations. Modelling provides a practical style of reasoning for economists.
Part II examines the process of model making. Economists give form to ideas, increasing formalization over time. Models provide reasoning resources to think through assumptions and implications.
Models can represent the world or be worlds in themselves. Case studies of specific models illuminate how modelling shapes beliefs about the economic world.
Overall, the book analyzes how mathematical modelling transformed economics into a field based on specialized tools and objects of thought - economic models. It seeks to understand economics from the perspective of economists at work with their models.

Here is a summary of the key points from the provided text:

The practice of economics shifted from formulating laws and theories in words to developing mathematical and graphical models. This marked a naturalization of modelling in economics.
Modelling represents a distinct style of practical reasoning. It involves giving form to aspects of the world using objects like equations, diagrams, or physical machines. These models then serve as tools for reasoning about the target system.
Models come in various forms, from theoretical constructs to physical machines. They provide different types of reasoning resources like simulation, metaphor, and experimentation.
In modeling as a method of inquiry, the world is represented or mirrored in the model. But models also act back on the world when used as tools to intervene or design policies. This reflects both the world in the model and models of the world.
The case studies in later chapters will illustrate how specific economic models were developed and used. They also discuss the philosophical issues around modeling as a way of doing and developing economic science. The goal is to understand modeling as its own autonomous epistemic genre of scientific practice.
In the past, political economy was primarily a verbal science based on describing general principles and laws of how the economy operated.
Starting in the late 19th century, economics became more technocratic and tool-based, relying heavily on models - mathematical, statistical, graphical or physical objects that can be manipulated.
Models are now ubiquitous in economics research, teaching, policymaking and more. They have become the primary way economists conceptualize and reason about economic phenomena.
One early pioneering economic model was François Quesnay’s Tableau Économique from the 1750s, which represented sectors of the French economy in a zig-zag diagram. Quesnay experimented with it to explore how the economy could grow.
Models raise philosophical questions about how economists create them, reason with them, and how they provide insight into the real world. Understanding specific early models like Quesnay’s also requires historical context that may not be obvious to present-day economists.

The passage summarizes the making, using, and learning from models as a way of doing science in economics in the following ways:

It describes how modeling gradually became a natural way for economists to do their work over the past 200 years, even though we don’t fully understand how this transition occurred.
It outlines three key periods - a prehistory with a few isolated models in the late 1700s/early 1800s, a first generation of modelers in the late 1800s who consciously used and justified models, and widespread modeling after the mid-1900s.
It provides examples of early models like Quesnay’s Tableau Economique, Ricardo’s farm accounting table, and Von Thunen’s farming diagram to illustrate the prehistory.
It discusses pioneers of the late 1800s like Marshall, Edgeworth, and Fisher who began regularly using models but had to justify this new approach, exemplified through their specific diagrams/mechanisms.
It characterizes modeling as a cognitive process that economists had to learn over time, going from reasoning verbally to with objects, in order to gain knowledge about economics. Understanding how specific models were built and used is key to this historical and philosophical change.

So in summary, it characterizes the transition to modeling as a gradual, complex process involving elements of both cognition and imagination, which economists had to consciously work through over the past 200+ years to make models a natural part of their scientific work. Understanding specific pioneering models is important for analyzing both the history and philosophy of this change.

This passage discusses the history and development of economic models from the late 19th century to the 1930s.
Some of the earliest model-makers were Edgeworth, Marshall, and Walras in the late 19th century. They developed graphical and mathematical representations of economic ideas and systems.
In the 1930s, economists began explicitly using the term “model” and conceptualizing their analytical tools as such. Important figures were Frisch, who developed one of the first mathematical business cycle models in 1933, and Tinbergen, who built the first large-scale econometric model of a national economy in 1936-1937.
Tinbergen helped transfer the term “model” from physics to describe these new statistical and mathematical representations in economics.
In the 1930s, others like Hicks, Meade, and Samuelson developed influential “Keynesian models” to explain and compare Keynes’ new macroeconomic theory.
By the 1940s and 1950s, modeling had become the normal way for economists to analyze and present economic ideas, relationships, and systems. The historical development traced models evolving from isolated examples to becoming a mainstream modeling practice.
The passage discusses the introduction of economic modeling as a new way of reasoning in economics. It compares early material models used in fields like astronomy and architecture to modern mathematical models used in economics.
Key points made:
- Early models were physical objects that could be manipulated to explore unseen workings, like planetary models. Economic models introduced a similar style of reasoning through manipulation.
- Models must be small-scale representations that can be analyzed and manipulated to indirectly inquire into the larger system/economy.
- Modeling became the preferred and accepted method of reasoning in economics by the mid-20th century. Disciplinary arguments came to rely on modeling abilities.
Crombie identified modeling as one of six styles of scientific reasoning. It grew from desires to imitate and understand nature through small-scale analogies. While analogies are less important now, the basic idea of representing a system to indirectly understand it remains key to modeling.
Introducing modeling brought a new way of reasoning to economics as a science - it became the “right way to reason” through building and manipulating representations.
Modelling emerged as a distinct style of practical reasoning in economics in the late 19th century alongside the use of mathematical techniques.
Early model-makers like Fisher developed hypothetical models using mathematical representations to explore ideas like general equilibrium, while others like Walras took a more axiomatic mathematical approach.
Modelling grew over time as it integrated with other styles like statistics and experimentation. It is now the dominant mode of reasoning in economics.
Adopting a new reasoning style like modelling requires arguments for its usefulness and results in changes to the science’s content and disputes over methodology.
Once accepted, a style becomes “natural” and seems the only objective way to reason within that domain, even though it has a contingent historical origin. Modelling is now so embedded it is hard to think of economics without models.
To understand modelling’s history, we must look at the models themselves rather than just the use of mathematics. Models are contingent products of their time and place.

The passage discusses the importance of understanding the development of modeling as an epistemic genre in economics. To do so, we need to analyze both the specific historical models (e.g. from Quesnay and Ricardo) as well as understand modeling in a more philosophical sense.

The key aspects of modeling discussed are:

Giving form to ideas - Models make vague economic ideas more explicit by representing them formally through diagrams, equations, pictures, etc. This can be thought of as a process of recipe making, visualizing, idealizing, or choosing analogies.
Making models rule-bound - In addition to giving form to ideas, models are made subject to rules of conduct and manipulation. Giving ideas a formal representation and making them rule-bound go hand in hand in the modeling process.

Some examples of how specific models gave form to ideas are discussed, such as Ricardo integrating accounting tables into a model farm, and Edgeworth visually depicting bargaining between Crusoe and Friday. The process of modeling involves both imagination to hypothesize how the economy works, as well as skill in representation.

The passage then transitions to discussing in more detail how economists make models by giving form to ideas and establishing rules for reasoning.

Here is a summary of the key points about modelling as a method of inquiry from the passage:

Modelling involves forming some kind of representation of something in the economy, though the nature of this representation can be understood in different ways (depicting, imagining, etc.).
Economists act in the process of model formation, creating models for particular purposes. Models are not self-generated.
Giving form to models can be described in different ways, such as using idealization, recognizing similarities, or creative analogy-building. There is no single logical process.
Model-making involves intuitive, imaginative and creative skills on the part of the economist. It is a skilled activity.
Economists acquire model-making skills through apprenticeship and experience. Different economists exhibit different combinations of articulation, construction and other talents.
Qualities of the final model, like being “fruitful”, indicate the model-maker’s talents were effectively applied. Models reflect the skills and abilities of their creators.
Modelling is a flexible method that takes diverse forms, as evidenced by the variety of examples discussed in the passage.
Representation involves “denoting” rather than accurately “representing” - models stand in a symbolic relationship to the economic system, not one of resemblance.
Representations become models when they have “resources for manipulation” - elements that can be used to reason with and explore the behavior of the system.
Reasoning with a model follows rules from both its format/language and economic content. The format provides formal rules, while the content provides economics-based rules. Together these determine valid reasoning.
Early models like Quesnay’s Tableau Economique had numbers that could be manipulated to reason about and explore the system’s behavior. Later diagrams like Frisch’s lacked these “reasoning resources” and could only support verbal description and reasoning.
Formalization involves giving economic ideas a particular form as a model, which then becomes rule-bound due to the formal rules of its language/format and the semantic rules of its economic content. This establishes the valid forms of reasoning possible with that specific model.

In summary, it discusses the key issues of representation, formalization of models, and how the reasoning resources and rules within models determine how they can be manipulated and used to explore economic systems.

Frisch developed early econometric models to study business cycles and macroeconomic dynamics. His first model was a simple schema or tableau, which had limited ability for quantitative or causal reasoning.
He then created a mathematical model connecting the annual production of capital goods (y) and consumer goods (x) over time (zt). This had equations that could be manipulated, allowing simulations.
Modeling represents ideas in a simplified format like equations or diagrams. Models have some representation of the real world but are smaller scale. They contain both economists’ existing knowledge and speculative theories.
Models can be investigated in their own right as “worlds in the model”. Economists experiment by changing assumptions and observing effects, using model rules. This helps understand the ideas and theories represented.
How investigations of models might also provide understanding of the real world is more complex. Model experiments are used to address questions, make changes, demonstrate answers deductively, and provide narrative explanations - acting as a method of inquiry.
Economic models embody ideas and knowledge about the economy, but are not themselves theories or data descriptions.
Models serve as intermediaries between the mathematics of economic theory and real-world empirical observations.
Models allow economists to think through and reason about combinations of economic concepts/relationships that would be difficult to conceptualize without a model. This is why economists create models.
Economists use models to experimentally explore their theories and intuitions. By manipulating parameters in a model, they can better understand implications, limits, and new hypotheses suggested by existing theories.
Models act as small worlds that economists investigate to answer questions. But philosophers are skeptical of how representative these small worlds are of the real world.
Modeling involves using the model world as an experimental object to represent parts of the real economic world and investigate how they behave. The question is then how to infer real-world implications from model experiments.
Model experiments provide insights but have limitations compared to real-world experiments due to using representations rather than the world itself. Economists make informal inferences by comparing model and real-world behaviors.
Figure 1.7 shows a diagram of the Phillips-Newlyn Hydraulic Machine, which was a large-scale physical model used to study macroeconomic concepts like money supply, income, investment, etc. Water flowing through the machine represented the flow of money in an economy.
The machine allowed economists to experimentally manipulate variables like money supply and measure the effects on things like income. This allowed testing of economic theories.
It demonstrates how models can be used as a method of inquiry, allowing probing of questions, experimental manipulations, and informal inference to develop explanations.
Modeling allows economists to investigate both their theories/ideas about the world (the world in the model) and gain insights into the real economic world by using the model to reflect on it.
Modeling has a long history as a method dating back to Galileo and remains important in natural sciences today. However, economists are unsure of the scientific credibility of modeling due to models being simple compared to the real world.
References are provided for further related works by authors like Morgan, Guala, Mayo, etc. that discuss modeling and inference from experiments.

Here is a summary of the provided sources:

The sources discuss the history and philosophy of using models as a method of inquiry in areas like economics, science, and the social sciences.
Many of the sources examine specific economic models and theories from thinkers like Quesnay, Irving Fisher, John Maynard Keynes, and others to understand how models were developed and used.
Other sources provide overviews of modelling as a method, addressing topics like idealization, abstraction, formalization, the relationship between models and reality, and how models are used as tools for learning, discovery, and understanding.
Methodological issues around models are discussed, like whether they should be seen as representations of reality, as tools for isolating specific factors, or as parts of distributed cognitive systems.
The history of modelling across different domains and time periods is explored, from ancient Chinese mathematics to modern computational economics.
Key figures in the philosophy of science like Kuhn, Goodman, Cartwright, and Morgan provide framings for understanding the role and epistemology of models.
In summary, the sources cover a wide range of historical and methodological issues regarding the use and analysis of models across different fields of inquiry.

Here is a summary of the article:

David Ricardo is considered one of the first “modern” economists due to his use of abstract, idealized examples in economic argument. However, Ricardo found these examples useful for addressing practical problems of his time. This suggests he was pioneering the use of economic modeling.
A famous example is Ricardo’s argument for free trade based on comparative advantage. He used a numerical example of trade between Portugal and England to demonstrate how specialization according to comparative advantage benefits both countries, even if one has an absolute advantage in both goods.
The article examines Ricardo’s numerical model from his Principles of Political Economy involving the distribution of output on a model farm. It analyzes how this model integrates various economic ideas and aligns with Ricardo’s experimental farming interests.
Ricardo’s model farm drew on contemporary ideas from experimental agriculture and can be viewed as modeling an individual farm, the agricultural sector, or incorporating different types of farming methods into one model.
Model-making for Ricardo involved selecting relevant ingredients or elements from economic theory and practice, and integrating them through reasoning to address practical problems. This demonstrated new possibilities for economic argument compared to prior models.

So in summary, the article analyzes Ricardo’s use of an early economic model and how it combined various elements and ideas to address agricultural issues, establishing Ricardo as an early pioneer of economic modeling.

Ricardo used numerical examples and models in his economic writings, even though his writing style was verbose and hard to follow for modern economists used to clear modeling.
His numerical examples and models were different from how economists nowadays think of models. For classical economists like Ricardo, the economy was governed by strict laws, not models.
One important numerical example Ricardo used was adding teams of 10 laborers at a time to cultivate a field. This became the basis for demonstrating his famous laws of distribution.
By integrating different numerical examples, Ricardo built up what can be seen as a “model farm.” He used this model farm to reason through and figure out the laws governing the economic system, through a process like “model farming.”
These numerical examples did more than just illustrate Ricardo’s arguments - they functioned to demonstrate his laws of distribution. The process of developing these models helped Ricardo create understanding and learn from the unexpected results.
To properly understand Ricardo’s numerical examples requires knowledge of the economics and economy of his time, which the passage provides through details about Ricardo and the agricultural debates of his era.
The passage provides context about Ricardo’s understanding of and engagement with agriculture, despite perceptions that he only knew finance and money as an economist.
It describes Ricardo’s country estate in Gloucestershire, which included agricultural lands and a manufacturing village. He took on responsibilities expected of a landowner, like supporting local infrastructure.
Ricardo had begun studying agricultural issues and the Corn Laws restricting grain imports by 1811. He opposed protectionist policies and read reports on the issue.
One report he discussed with Malthus contained witnesses’ statements and farm accounting data. His future land agent Edward Wakefield was among the witnesses.
Wakefield later kept Ricardo informed about estate management, tenants, and crop prices as his land agent.
Serving in Parliament from 1819, Ricardo gained further knowledge of contemporary agricultural challenges through select committee investigations.
This experience informed his economic writings, speeches, and policy positions regarding agriculture and related issues population growth and food prices.
Experimental farming was important in late 18th and early 19th century Britain to improve agricultural productivity and output. Experiments looked at animal husbandry, fertilizers, cultivation methods, machinery, etc.
Reports of experiments by Arthur Young, William Marshall, and others were published and circulated to advise others.
Wealthy landowners established experimental farms and held agricultural shows to disseminate new practices, like at Holkham Hall and Woburn estates.
There was political involvement in agriculture as the genteel pursued improving farming. Agricultural societies also promoted experimentation.
Humphry Davy’s lectures through the Board of Agriculture from 1803-1812 brought a scientific approach. But practical experimentation was still important.
Ricardo’s political economy concepts drew on the empirical findings of contemporary agricultural experiments in Britain. The experiments provided subject matter and his numerical models paralleled experimental reports.
Ricardo developed numerical examples and accounting methods to demonstrate his economic arguments about growth, distribution, and other topics. This went beyond just using tables.
He was familiar with contemporary experimental farming reports from major landowners and publications like the Farmers’ Journal. These reported on experiments, costs, profits, etc.
Ricardo’s numerical examples in his Principles of Political Economy mirrored the format and content of real farming reports from his time. They discussed problems in agriculture and analyzed factors like rent, capital, profits.
His “model farm” examples integrated his economic ideas with knowledge of agricultural practices. Through his accounting methods on this model farm, he formulated laws of distribution.
The numerical accounts served as reasoning tools complementing his verbal arguments. Each example enabled working through concrete scenarios in his model farm economy. So they played a special role beyond just illustration.

In summary, Ricardo blended knowledge of actual experimental farming with economic analysis through his use of numerical accounting methods on a conceptual “model farm”. This integration helped him develop and demonstrate theories of economic growth, distribution, and other topics.

This passage discusses David Ricardo’s use of numerical experiments and model farming in his work Principles of Political Economy and Taxation. Some key points:

Ricardo aimed to understand the laws governing the distribution of gains between economic classes (rent to landlords, profits to capitalists, wages to laborers).
He used numerical experiments/accounting to clearly demonstrate how rent arises and is determined, and how changes in elements like agricultural investment affect this distribution over time.
The first accounting shown uses a hypothetical model farm with three plots of land of different fertilities. It shows how cultivating lower quality land leads to rent on the higher quality land.
These experiments assume classical economics principles like equalization of profits and profits determined on least productive land.
They allowed Ricardo to integrate different factors into a model farm and gain unexpected insights into the small-world economy and how it may respond to changes.
Later chapters will discuss how such model building and experimentation can generate unexpected results by adopting a new formal mode of reasoning, and how experimentation is a general feature of model functioning.

So in summary, it discusses Ricardo’s use of numerical experiments and model building to clearly demonstrate the dynamics of economic distributions and gain insights, building on classical assumptions through integrated and varying scenarios.

This passage provides context around Ricardo’s third numerical accounting example from Chapter II of his book Principles. Some key points:

Ricardo discusses technical improvements in agriculture, namely introducing a rotation of turnips or using a more “invigorating manure.”
He presents a numerical example to show how these improvements could increase agricultural output from the same land, thereby lowering rents. The example shows increasing capital investments (portions 1-4) and the resulting outputs both before and after the technical improvement.
This numerical experiment Ricardo conducts models how technical change on his hypothetical “model farm” could affect both the amount of capital needed in agriculture and the level of rent.
The passage then provides a documented example from a 1817 issue of the Farmers’ Journal of an actual field experiment conducted near Ricardo’s estate, comparing potato yields from plots with varying amounts of manure. This real-world experiment supports the type of technical improvement - use of manure - that Ricardo incorporates into his numerical modeling.
Overall, the passage provides context around Ricardo’s use of numerical examples and hypothetical modeling to explore how technical improvements in agriculture could impact rents, showing he was engaging with real agricultural experiments and issues of the time.
David Ricardo conducted numerical experiments in political economy similar to agricultural experiments of his time, using accounting formats.
One key experiment added labor to a field in units of 10 men. It included prices of corn and rent in monetary terms for the first time. As more labor was added, output increased but at a declining rate, and rent increased in both corn and monetary terms.
This fifth experiment explained Ricardo’s statement that landlords obtain a greater share both in terms of corn and monetary value as labor increases.
Ricardo later expanded on this experiment to examine the effects of increasing labor on wages and profits, integrating these factors into his model farm.
This fully integrated model farm showed how profits decline and rent takes a greater share as more labor is employed. It demonstrated Ricardo’s laws of distribution.
The model farm revealed that adding infinite labor would cause profits to fall to zero, setting a stagnant base level of zero profits for the entire economy and halting further capital investment and growth. It was an effective tool for deriving complex results about the economy.

Here is a summary of how events might unfold:

As more labor is added to farming through the adoption of spade husbandry techniques, yields per acre would initially increase due to improved soil cultivation methods. This would allow farm output and profits to rise.
However, according to Ricardo’s economic model, adding more labor would eventually drive down farm profits. As profits fall, farmers would have less incentive to invest in their businesses or adopt new technologies.
Over time, declining profits could lead to stagnation in the agricultural sector as investment dried up. This would negatively impact the broader economy.
Falling farm profits would also squeeze the returns to landlords and capital. The economic surplus from farming would be increasingly appropriated by laborers in the form of higher wages.
As more people were employed in low-profit farming, population growth would continue to be supported. But the long-term result, according to Malthusian theories, would be a subsistence level standard of living for the lower classes.
Without technological progress or a transition to more productive industries, the economy could become trapped in a cycle of low wages, minimal capital investment and stagnant growth. This was the potential dystopian outcome envisioned by both Ricardo and Malthus.
The spade husbandry debate highlighted these risks while also attempting to address immediate employment and poverty issues through labor-intensive agriculture. It provided a real-world example to illuminate Ricardo’s theoretical model.

Here is a summary of the extract:

The extract provides details about Ricardo’s use of a model farm to investigate principles of political economy related to rent, profits, and population growth. Ricardo developed numerical accounts for an imaginary farm over multiple chapters, allowing the model farm and its behavior to emerge gradually as he posed different possibilities.

Key points:

The model farm represented and functioned in three ways simultaneously: as a model according to economic principles, as a model of real farms through using contemporary numbers, and as a model of the entire farming sector by showing aggregate-level effects.
It incorporated Ricardo’s definitions, concepts and assumptions from classical political economy, such as rent, profit rates, and factors of landowners, capitalists and laborers.
Through the model farm accounts, Ricardo was able to formalize and integrate both agricultural facts and his theoretical ideas about distribution of rents, profits, etc. according to the laws of classical economics.
The model farm had an independent existence that allowed it to autonomously demonstrate behavior consistent with Ricardo’s classical economic system, though it was not the system itself.

In summary, the extract discusses how Ricardo developed his model farm accounts to both represent specific concepts from classical economics and real farming practices, integrating theory and data to investigate questions about rents, profits and population within a single numerical modeling framework.

Ricardo’s model farm accounts embody key elements of classical economics, like the labor theory of value, the tendency of the profit rate to fall, etc.
These elements are not in one-to-one correspondence with the model farm accounts, but are gradually embedded in them as the accounts build up through different scenarios.
The numbers and format of Ricardo’s accounts were based on realistic data from farms of the time and experimental farming reports, showing his model was grounded in empirical facts, not purely abstract.
The scenarios Ricardo modeled, like population growth and adding more laborers, captured important economic issues of the day like those debated with Malthus.
Ricardo’s model farm functioned not just as an individual farm, but represented the agricultural sector as a whole, as indicated by price changes from individual actions. It was effectively a “giant aggregate farm” standing in for the overall economy.

So in summary, while an abstract model, Ricardo’s farm accounts incorporated key classical assumptions, were empirically grounded, addressed contemporary issues, and functioned as a model of both individual farms and the entire agricultural sector or economy.

Ricardo’s model farm accounts serve both to represent individual farms and the aggregate economy. However, aggregating in this way creates issues of interpretation.
While the numbers in the model farm tables relate to individual farms with varying outputs, Ricardo moves from analyzing a single labor input level to attributing it to the whole economic system, raising questions about how patterns carry over across different input levels.
Ricardo’s model farm brought together both conceptual elements (definitions, laws) and empirical elements (numbers from farming) like ingredients in a new recipe. However, his model farm did not strictly follow real farming accounts - it operated according to classical economic laws and concepts.
The model farm acted as a mediator between Ricardo’s ideas about the economy and real economic realities. Its independence of elements allowed it to investigate both theories and the world.
Integration of elements in the model is important for its usefulness. For Ricardo, accounting conventions provided the integrating element that molded the different parts together.
Using just words, Ricardo and Malthus struggled to reason through complex issues involving multiple variables and aspects. The model farm accounts provided a way to integrate different concepts and analyze their interactions, overcoming limitations of two-dimensional verbal/tabular reasoning.

Here is a summary of the key points about Ricardo and his model of agriculture:

Ricardo used pen-and-paper experiments with a model farm to test and illustrate his economic ideas, not just to show individual elements but how they work together in varying combinations.
His model farm represented both an individual farm and the agricultural economy as a whole. This allowed him to show multiple variables changing at once.
The model farm formalized his ideas and made them behave systematically according to set rules, not arbitrarily. When one variable changed, others had to change consistently.
He used accounting methods to integrate all the elements of the model farm so they worked together consistently. Changes in one part rippled through the whole model.
The “laws of distribution” emerged deductively from experimenting with the model farm as an integrated whole, not just from assumptions alone or individual components. These laws dictated how wages, profits and rent are determined as the economy changes.
An early example was a table from his 1815 essay modeling how profits and rent varied with increasing capital inputs on less fertile land, keeping the profit rate constant across land. This illustrated interdependence of variables.

So in summary, Ricardo’s model farm allowed him to systematically test relationships between economic variables and deduce general principles of how the economy operates.

Ricardo was interested in modeling economic growth and capital accumulation over successive time periods, not just alternative scenarios. Time was necessary to model a growing population and how agricultural profits may fall over time.
Ricardo’s 1815 table demonstrated something novel - that increased capital inputs could initially raise profits but eventually reduce them, while rent and net produce continued rising. This surprising result alerted Ricardo to new insights from numerical modeling.
The table was intended to illustrate theoretical principles, not provide accurate empirical data, as Ricardo acknowledged with a footnote disclaimer. However, there is reason to doubt how far the numbers departed from economic realities given Ricardo’s knowledge and experience.
The table reflects classical economists’ concerns about feeding a growing population while worrying over the “tendency of the profit rate to fall” with accumulation of capital over successive time periods.
Ricardo took the table’s findings of initially rising then falling profits to be an essential element of his argument, not just an artifact of the specific numbers used. It demonstrated to him the potential of numerical models to generate unexpected insights.

So in summary, the key points are around Ricardo’s use of modeling over time rather than just scenarios, the novel insights generated from the 1815 table, and the theoretical versus empirical aims of the modeling.

The article examines the development of modeling and mathematization in economics in the late 19th century, using the Edgeworth Box diagram as a case study.
It argues that mathematization and modeling are related but not the same. Modeling involves developing a representation of the world, while mathematization refers specifically to using mathematical language.
Early developers of mathematical economics argued it would make the field more scientific by allowing for more exact and rigorous expression and reasoning. However, mathematization was controversial within economics.
Modeling gives formal rules for reasoning about ideas of the world, while the content can be expressed in non-mathematical languages as well through analogies, verbal descriptions, etc.
Developing new forms of expression through modeling and mathematics prompted not just changes in how economists worked, but conceptual changes in what they expressed. It allowed for new economic ideas to emerge.
The transition from verbal to mathematical expression should not be seen simply as translation between languages. It involved a complex process of visualizing the economic world in a new way and developing images to represent it.
The history of the Edgeworth Box diagram is used to illustrate how in practice, cognition, visualization, and conceptual development were interlinked in the development of mathematical modeling.

The passage discusses the process of mathematization in economics. Some key points:

Translating economics into mathematics is more complex than simply translating words into equations. Economists must choose an appropriate mathematical language/formalism to represent economic ideas and behaviors.
Different mathematical representations (equations vs diagrams for example) have different semantics, syntax, and implications for how the models can be manipulated. They are not perfectly equivalent.
Economists are not merely transcribing pre-existing mathematical laws of economics. They are using their intuitions and imaginations to represent and explore opaque economic concepts through model-building.
Model-building helps economists express and understand their ideas through creating simplified representations/images of the economic world. It involves both conceptual and perceptual work.
Mathematization involved economists imagining new mathematical representations/models of the economic world that could accurately portray their ideas in precise formal languages, just as the verbal description of economics developed over time.
It was a complex historical process of interaction between economists and mathematicians, where the image and content of both fields evolved together, not a simple translation. Over time, new mathematical ways of representing the economy became taken for granted.

So in summary, mathematization in economics involved much more than simple translation - it was a creative process of building new conceptual and mathematical models or “versions” of the economic world.

The passage discusses the process of creating economic models like the Edgeworth Box and how this relates to visualization and cognition.
It compares an artist’s visualization of how the Edgeworth Box model was developed to the original diagrams created by economists and some modern representations.
The artist’s depiction portrays the process as starting with the whole world, then simplifying, isolating, and abstracting to eventually arrive at the Edgeworth Box model.
This echoes descriptions by philosophers of science of how mathematical modeling involves simplification, isolation, abstraction, and idealization.
However, the passage notes this view overlooks a crucial aspect - that modeling also involves creating new representations and conceptual elements that did not previously exist in verbal or other forms.
Creating these new representations changed how economists visualize and understand the economic world by learning to think in the new representational forms.

So in summary, it analyzes different perspectives on developing the Edgeworth Box model and discusses how modeling both simplifies real-world complexity but also creates novel conceptual and representational elements that changed economic thinking.

Here is a summary of the key points about the economy:

The passage describes how economists simplify the complex real world to build economic models and concepts. It starts with a basic example of two individuals bartering goods, and builds up more complex concepts from there.
An artist created illustrations showing the step-by-step simplification from detailed characters and goods to an abstract representation of bartering between two individuals. This helps represent the simplification process economists use.
Economists take the simplified representation and add further conceptual elements like indifference curves to model preferences and build tools like offer curves, contract curves, etc. This enables reasoning about concepts like Pareto optimality and Nash equilibrium.
The artist can only take the simplification process so far through illustrations. Economists must use additional conceptual/theoretical elements to analyze the economy and develop economic theories.
The illustrations depict a perceptual space, while economists work in a conceptual space using theoretical tools beyond just the perceptual representations. This allows them to analyze and make arguments about the economy.
In summary, it describes how economists simplify the real world to develop basic representations, then layer on additional conceptual elements from economic theory to enable complex analysis and theoretical insights.
The Edgeworth Box diagram was first introduced by economist Francis Ysidro Edgeworth in his 1881 book Mathematical Psychics. It represents the exchange of utility between two individuals trading two goods.
Edgeworth viewed mathematics both as a language for expressing economic ideas precisely, and as a tool/instrument for deductive reasoning about economics. But he also saw it as an instrument of imagination to capture unseen economic behaviors.
Subsequent economists like Pareto, Bowley, and Leontief built on Edgeworth’s initial diagram, gradually adding more complexity to represent more goods, individuals, constraints, etc. This process helped develop the conceptual foundations of the modern Edgeworth Box model.
Reconstructing how economists originally imagined and created the Box diagram is important, as it addresses how they developed clarity around dimly perceived concepts, unlike later economists who already knew the fully developed model.
Looking at the actual historical sequences of diagrams, rather than generic accounts of model development, provides insights into how economists gradually built up the conceptual world contained within the Edgeworth Box model through their successive diagrams.
Edgeworth visualized economic concepts like exchange mathematically using diagrams and symbols, similar to how mathematicians think and reason with imagined worlds.
His most famous contribution was developing the “Edgeworth Box” diagram to illustrate bilateral exchange between two individuals.
The diagram represented the two goods being exchanged on the x and y axes. It showed indifference curves for each individual’s utility from different combinations of goods.
The “contract curve” represented the set of possible exchanges that benefit both parties. This region was constrained to lie between the individuals’ indifference curves.
Prior diagrams by Jevons and Marshall represented concepts like utility or trade in different ways. Edgeworth’s diagram built on their ideas while specifically illustrating bilateral exchange between two agents.
His mathematical and diagrammatic representation helped establish economics as a rigorous analytical science rather than just analogical thinking, though he still used some physical analogies initially.
Edgeworth created a diagram known as the Edgeworth Box to represent the bargaining process between two individuals, Robinson Crusoe and Friday, who are trading goods with each other on a deserted island.
The diagram shows the “contract curve” where trades make both individuals better off. As they bargain from the origin point, they can move along this curve to reach mutually beneficial agreements.
The exact point they settle on depends on their relative bargaining strengths. The diagram demonstrates Edgeworth’s concepts of indeterminate contracts and how arbitration could help resolve deadlocks.
Edgeworth then developed the diagram further to consider what happens as more traders enter the market, representing imperfect competition. He was able to gain new insights by reasoning through the bargaining and agreement process using the diagram.
Edgeworth saw the diagram as providing an “abstract typical representation” of exchange that allowed him to establish general principles through reasoning about a specific, representative case using a form of “mathematical induction.”
This highlights how economists can develop understanding by creating models to represent hypothetical situations and reasoning through the implications, even if based on a single case.
The passage discusses Vilfredo Pareto’s use of diagrams to develop economic arguments, specifically relating to exchange between individuals.
Pareto uses a series of diagrams as the primary means of mathematical reasoning in the text. He includes a more general algebraic/calculus treatment in an appendix.
Like Edgeworth, Pareto introduces specific diagrams with general arguments, but unlike Edgeworth the arguments are not analogical. He relies more extensively on the diagrammatic form to develop his arguments.
Pareto portrays individuals seeking their best outcome but facing constraints/obstacles. They must take different paths around these obstacles through trial and error, reflected in Pareto’s diagrams.
Two key diagrams that advance Edgeworth’s work are Figure 16 creating a “box” form and Figure 50 developing the concept of a “Pareto improvement”.
Figure 50 shows indifference curves for two individuals and defines a Pareto optimum. It establishes the possibility for collective utility gains through gains to both parties.
For Pareto, the essential reasoning is done diagrammatically, not just off the diagram. The diagrams represent his mathematical economic world used to build arguments.
Despite its small scale, the box diagram reaches fundamental results like Pareto optimality and the First Welfare Theorem due to its flexibility across different situations.
The Edgeworth Box model originated from Edgeworth’s 1881 diagram, which was not actually a box but an open plane. It depicted two individuals lining up side by side to exchange quantities along one axis, rather than facing off at opposite corners of a box.
Pareto in 1906 was the first to depict the individuals at opposite corners of a rectangle, representing fixed quantities of goods. This became the standard orientation of the Edgeworth Box model.
Important innovations include Lenoir and Bowley moving the starting point for exchange (“endowment point”) inside the box, rather than each individual starting with their full endowment as in Edgeworth’s model. This allowed indifference curves to be drawn fully inside the box.
Depicting the initial endowment point inside the box influenced later neoclassical economics to focus on efficiency given initial endowments, rather than addressing questions of initial wealth distribution and equity.
The model was originally developed to analyze bargaining between two individuals (Crusoe and Friday), not market exchange with many traders, as Edgeworth intended it to analyze. Later contributors used it to represent different economic scenarios.
Details like whether the axes meet or extend beyond the box were meaningful aspects considered by different contributors to the development of the model.
The article discusses the development and evolution of the Edgeworth Box diagram over time, from its origins to its more standardized modern form.
Early diagrams by Edgeworth, Pareto, and others were more flexible in their representation but more economical. Later diagrams became more standardized and complex as the model was used to represent more concepts.
A key point is how the box represents resource endowments - a fixed size box represents a fixed total resource situation, while changing the box size represents changing total resources. This was an imaginative way to model resource flexibility.
Over time, indifference curves and contract curves were added to the diagrams to demonstrate exchange and welfare. Individuals trading inside the box were represented only symbolically through indifference maps and endowment points rather than as distinct people.
The standardization of the box diagram after the 1950s made it harder to represent some of the original imaginative variations, but it also allowed the model to be applied to new economic domains like production and trade.
The evolution of the box diagram involved economists using their imagination to visually represent exchange and gradually building up the analytical apparatus contained within the model through new diagrammatic representations.
Mathematical models in economics, starting with the Edgeworth box diagram, introduced new conceptual elements and apparatus that did not previously exist and could not be fully expressed verbally or mathematically without the diagrams.
Reasoning with these diagrams depends not just on mathematics but also on understanding the new economic concepts represented in the diagrams. The diagrams allow for new types of analysis and lines of reasoning.
Diagrams may have an epistemic advantage over verbal or algebraic representations because they express spatial and relational aspects of problems rather than just temporal/logical relations. This enabled creative new modeling approaches.
The form of representation, whether sentences, algebra, or diagrams, dictates the types of operations and reasoning that can be applied. Diagrams allowed for different and more efficient reasoning about economic concepts like exchange equilibrium.
Examples from the history of economics show how diagrammatic models introduced novel conceptualizations that changed the substance of economic theories in productive ways.
Models play an important role in the mathematization of economics. Economists cannot fully mathematize the economic world all at once, they need models to develop the new vocabulary and ways of thinking.
The Edgeworth box diagram provides an example of how this process occurred. Developing this model allowed economists to visualize and imagine exchange relations in a new way. It helped create new concepts like indifference curves.
Creating models like the Edgeworth box generates conceptual content that becomes foundational for mathematical economics. Concepts developed in specific models can travel to other models and applications.
Models represent something independently of verbal accounts using different concepts and arguments. They allow the representation of something not easily expressed in words.
This quality of models being independently conceptual means they became constitutive, not just illustrative, of modern economics. They formed important building blocks for developing a mathematical version of the economic world.

So in summary, the key point is that models played a core role in the mathematization of economics by allowing economists to imagine, visualize and develop the conceptual content and foundations of mathematical representations of the economic world.

The development of general equilibrium theory provides a parallel conceptual account alongside mathematical developments in economics in the mid-20th century. Some historians of economics note the emergence of small-scale modelling in general equilibrium theory at this time, which they interpret as economists copying physics’ use of modelling.

However, this account suggests a different chronology, arguing that models played an important role in the mathematization process in economics much earlier. Small-scale modelling in general equilibrium theory in the mid-20th century is viewed not so much as economists copying physics, but rather as a natural extension of the increasing use of models earlier in economics’ mathematical development.

So in summary, it presents an alternative view that challenges the idea that modelling was copied from physics in the mid-20th century, and instead argues models had a more significant role throughout the mathematical development of economics, not just in the mid-20th century moment of general equilibrium theory. It offers a reinterpretation that gives models a more prominent conceptual place earlier in the mathematization process in economics.

Here is a summary of the key points from the passage:

Early classical economists like Adam Smith provided more nuanced, “thick” descriptions of human behavior and motivation rather than simplified models. Smith’s account was too complex to serve as a useful analytical model.
Thomas Malthus was one of the first to construct a simplified model of “economic man” with just two key motives - self-interest and desire for offspring. He hypothesized how the interaction of these motives and two population laws would lead to economic cycles among the poor.
Malthus’ model was thin enough to reason with analytically, and he explored counterfactual scenarios by considering how outcomes would differ if his model man acted differently. This was one of the earliest uses of an explicit economic model of human behavior.
Later economists increasingly abstracted their characterizations of “economic man” to create ideal types that were stripped down to just the motivations and behaviors relevant to economic analysis. This allowed for more mathematical and deductive reasoning but resulted in more caricature-like models.

So in summary, the passage traces the evolution from richer descriptions to increasingly simplified and idealized characterizations of “economic man” that served as useful analytical models for economists over the nineteenth century. Malthus was an early pioneer of this modeling approach.

Thomas Malthus and John Stuart Mill both developed portraits or models of “economic man” to help understand economic behavior and laws.
Malthus’ economic man was primarily driven by the motivation of sexual reproduction, which provided the basis for Malthus’ theories about population growth outstripping food supply.
Mill developed a leaner characterization of economic man, motivated solely by a desire for wealth and dislike of labor/love of luxuries. This allowed Mill to develop political economy as a more scientific discipline focused narrowly on economic motivations.
Later conceptions like Weber’s “ideal types” and Menger’s “human economy” moved toward more abstract conceptualizations. Weber saw ideal types as purified fictions constructed from experience to represent phenomena.
Menger’s human economy portrait focused on an individual premeditatively seeking to satisfy material needs given their personal nature and economic situation/constraints at a given time. This provided a basis for understanding economic choice and action.

So in summary, Malthus and Mill developed early portraits of economic man to understand behaviors and laws, while later conceptions like Weber and Menger moved toward more abstract conceptualizations and ideal types as tools for analysis and understanding economic phenomena.

Carl Menger proposed that individuals choose quantities of different goods in order to satisfy their needs in a particular order, with necessities coming before less important needs. He depicted this in a schedule showing an individual’s ratings of satisfaction from different goods.
Menger aimed to ascertain the simplest elements of human economy through a partially empirical analysis. He sought to build up concepts of typical economic behavior by defining the simplest elements, rather than testing if they exist empirically.
Menger’s “economic man” was an ideal type, which was neither idealized nor a type in the normal sense. It was a conceptual model used as a benchmark for understanding economic behavior, rather than a hypothesis or empirical description.
William Jevons took a similar approach of constructing an abstract conceptual model (“economic man”) but represented it symbolically using mathematics. This allowed the model to be manipulated and applied in new contexts in a more powerful way than verbal representations. Both Menger and Jevons developed abstract conceptual portraits of human economic behavior through a process of concept formation, though Jevons’ use of symbols allowed further development.
Jevons presents a new conception of economic man that moves away from Mill’s motivation of accumulating wealth, towards man seeking to maximize utility/pleasure from consuming goods.
Jevons mathematizes Bentham’s ideas about utility, reducing it to quantities that can be plotted on a graph to show how pleasure declines with consumption. This characterizes economic man’s behavior mathematically.
By portraying man as making calculations to maximize utility, Jevons implies people reason using the same mathematical principles. However, utility is actually subjective experiences internal to the individual.
Jevons considers economics to be properly expressed through mathematics. Others debate whether the “Book of Nature” for economics is truly written in math.
Regardless, Jevons’ mathematical portrayal significantly influenced later conceptions of economic man and tied models more closely to formalism using symbols rather than informal language.
Jevons thus helped establish modern economics by changing its language to the more exact but constrained forms of mathematics, with implications for how economists abstract human behavior.
Jevons introduced the concept of the “calculating man” or economic agent who calculates utilities and acts rationally to maximize pleasure and minimize pain.
Later economists like Edgeworth and Pareto further developed models using Jevons’ calculating man as an agent within the models.
A key difference from classical economics is that in marginalist theory, the actions of each individual agent matter, rather than just aggregate behavior. A change in one individual’s preferences could change market prices and outcomes.
Frank Knight argued that for the calculating man to properly function in neoclassical economic models, it had to be endowed with perfect information and foresight. This made the model more idealized and separated from real human behavior.
Knight referred to this idealized model as a “slot machine” - it had no intelligence, competed, bargained, or interacted with other humans. It was purely a utility-maximizing agent for use in mathematical models.
While useful for theoretical analysis, Knight argued this idealized economic man bore little resemblance to actual human economic behavior and could not be used for understanding or policy advice regarding the real economy.
Frank Knight created a specific caricature of “economic man” for neoclassical economic theory called “slot-machine man”.
This conception of economic man was given exaggerated qualities of perfect information and foresight that went beyond real human behavior. It portrayed economic actors as mechanistic automatons without judgment or intelligence.
Knight created this caricature through the addition of fictional/false assumptions, rather than simply abstracting away complicating factors as earlier economists had done.
Knight’s “slot-machine man” served as an idealized theoretical construct that allowed economists to explore economic behavior and theory in its most exaggerated form.
Another conception that gained prominence was “rational economic man”. This portrayed economic actors as rationally choosing between options to maximize outcomes. It depicted them as psychologically thin or two-dimensional.
“Rational economic man” emerged from stripping away the underlying psychology from earlier portraits by Jevons and Menger, and defining rationality specifically as rational choice behavior tied to neoclassical economics. It came to function as a model of rational behavior despite its cartoon-like qualities.
Weber and Robbins defined economics as dealing with scarcity and choices between ends and means. This shifted economics’ focus from material welfare to human behavior in allocating scarce resources.
This change made choices central to the conception of economic man rather than desires or needs. Rational economic man was defined by his rational choices rather than his underlying motives or feelings.
By focusing on rational choice, economists took personalities and psychology out of the model of economic man. He became more of a conceptual symbol to analyze outcomes rather than describe actions.
However, rational economic man was also seen as a normative model, guiding how people should behave rationally. The reasons for action came from rationalizing consequences rather than initial emotions.
This history shows how economists created increasingly abstract and idealized models of economic man to make economics more scientific by focusing on key elements of behavior. Different terms like simplifying, abstraction, idealization were used to describe these model-making processes.
But it is difficult to precisely define the modeling processes used by different economists and to generalize across their approaches. Philosophers of science also use idealization in different ways, complicating efforts to analyze these economic modeling techniques.

Here is a summary of the key points about processes scientists use in model-making based on the passage:

Model-making involves a creative process similar to how artists create caricatures or character portraits. It’s not just logical analysis.
Examples of early economic models like Knight’s “slot machine man” and Mill’s characterization of Scrooge showed how certain characteristics could be exaggerated to create insightful portraits/models.
Creating a good caricature or model involves both accurately portraying similarities to reality, but also exaggerating certain features in a way that provides new insights when recognized.
The process of creating a caricature, as shown through Philipon’s drawings of Louis Philippe as a pear, involves both losing features of reality while gaining features of the caricatured form. It’s a creative transformation, not just selection/addition of features.
No single concept of “idealization” fully captures this process, which involves generalization, subtraction, abstraction, addition, and exaggeration in a holistic, creative way to develop insightful models or portraits. The history of model-making is messier than a logical analysis suggests.
Model-making in economics can be understood as a process of caricaturing or idealizing - selecting, simplifying, and transforming elements from existing descriptive accounts of people or phenomena.
Scientists can only idealize from materials they already have - e.g. previous descriptions, observations, or existing models. They transform and simplify these existing accounts.
Economists have a substantial observational base of knowledge about economic behavior from observing themselves and others. They also have existing theories and models from previous economists.
Later models of “economic man” transformed and simplified earlier descriptive accounts or existing model versions. E.g. Jevons transformed Mill’s characterization into a calculating man.
The choices economists make in transforming and idealizing are shaped by the traditions and scientific practices of their time. Different traditions produce different characterizations of economic man.
Changes in economic man models parallel broader changes in economic methods, contents, and paradigms over time. Different eras caricaturize economic man in ways salient to analyze the economy of that period.
After being treated primarily as an observational sketch (phase 1) and theoretical construct (phase 2), economic man is now shifting back toward being a more empirically grounded descriptive account (phase 3) through processes of de-idealization.
Early economists like Mill portrayed economic man as seeking wealth, while Jevons viewed him as seeking to maximize pleasure from consumption. Menger presented him as satisfying needs through sensible choices.
These were simplifying models that focused only on the explicitly economic aspects of human behavior, reducing complexity. Each tried to represent real human behavior in a scientific way.
Jevons’ economic man marked a shift towards idealizing the models to endow them with more perfect economic qualities. Later models like Knight’s had extraordinary knowledge and certainty.
Recent developments have aimed to “de-idealize” the models by adding back complexity and limitations like bounded rationality to make them more descriptively accurate. New research uses experiments, simulations, and other methods.
This has led to more complex and “fatter” portraits of economic man that are less driven by theoretical requirements - depicting him as able to learn, bargain, act strategically, with memory and potential for happiness. Removing him from the “dismal science” portraits of early economists.

The passage discusses how concepts of “economic man” and models of rational economic behavior have evolved over time. Originally, economic man was seen primarily as an abstraction used in mathematical models. However, more recently models of economic behavior are being manipulated and tested in experimental settings as well. This allows models to be used to inquire into real world behaviors, though in radically new ways compared to the original conceptions. Through experimental testing and manipulation, models are once again becoming useful for understanding the actual world, not just theorizing within abstract models. The passage acknowledges influences on this work from earlier discussions of models and modeling. It provides context on the evolving roles and uses of models of economic behavior over the last century or so.

Here is a summary of key points about metaphors and analogical models in economics from the given text:

Economists often use metaphorical language to describe economic concepts, like referring to money as a “liquid.” Metaphors can provide the basis for substantive analogies.
Turning a metaphor into an analogical model involves both cognitive and imaginative work. The economist must choose another system/object to draw an analogy with based on some similarities to their beliefs about how the economy works.
In choosing an analogical world, economists place constraints on the form and content of the model. They can then develop the model using these constraints to explore implications and interpret the economy in those terms. This is a cognitive process of comparison and translation.
Filling in the details of the chosen analogical world presents questions that must be answered. Solving these allows economists to gain new understanding of the economic system.
A metaphor on its own is one-dimensional, but developing it into a model along various dimensions allows the scientist to gain more from invoking the metaphor. The choice of analogical world provides more explicit guidance for depicting the properties of the modelled economic world.
Some examples of analogical worlds economists have used include blood circulating in the body, water in natural ecosystems, and tidal flows between oceans and lagoons.
The Newlyn-Phillips Machine was a hydraulic machine built in 1949-1950 to represent the macroeconomy. It used colored water flowing through pipes and tanks to dynamically model economic concepts like national income, expenditure, consumption, savings, investment, etc.
The prototype (Mark I) was first demonstrated at the LSE in 1949 and was displayed at conferences to help teach economic theories visually. A more advanced Mark II version was later built.
As an analogical model, it mapped economic concepts onto physical components like tanks (to represent money pools) and valves/slides (to represent behavior of groups like consumers, investors). The flows of “money” represented circular flows in the macroeconomy.
While a photo shows the actual machine, the diagram helps illustrate how it worked as an analogical model - mapping economic relationships and theories onto a physical hydraulic system to dynamically demonstrate theories in action. It helped economists imagine and grasp complex relationships by representing them visually through a working physical model.

This passage provides context and details about the Mark II Machine, which was a physical model used to demonstrate economic relationships and how the national economy works. Some key points:

The Machine modeled the national economy as a system of water (representing money) flowing through pipes, tanks, and valves. This allowed it to show how different economic variables and policies interact dynamically over time.
Several versions of the Machine were built and sold to universities around the world for teaching purposes. They could represent different national economies and economic theories.
It functioned as a programmable analog computer, allowing users to set initial conditions and model equations governing the system. Outputs would be recorded on charts.
It was intended to make complex economic concepts accessible while also demonstrating them rigorously. Students could manipulate different parts of the Machine to experience policy coordination challenges.
Over time it became iconic, depicted in newspaper cartoons and articles. While playful, these conveyed the quirkiness of the real-world economic system it aimed to illustrate.

So in summary, the passage provides historical context about the development and uses of a physical macroeconomic model called the Mark II Machine, which creatively visualized national economies and policy interlinkages through an analogous hydraulic system.

Rowland Emett saw a political demonstration in London in the early 1950s that inspired his Punch cartoon of the Phillips Machine. Accounts differ on the exact date, with Emett remembering it as 1949 when Dalton was Chancellor, but the cartoon date of 1953 fits Rab Butler being Chancellor.
Walter Newlyn and Bill Phillips were the co-inventors of the Phillips Machine. They came from different backgrounds but shared talents.
Newlyn left school at 16 but worked his way up in a grain trading firm in London. This experience on the Baltic Exchange dealing with large commodity trades informed his understanding of money flows in the economy.
Phillips grew up on a farm in rural New Zealand in the early 20th century. His family developed innovative water and electricity systems using the stream on their property, which Phillips helped enhance.
The cartoon captured the dual nature of the Machine as both a working hydraulic device and a representation of a living economy. It brought attention and scrutiny to the Machine over subsequent decades.

Here is a summary of the key points about the ir beds:

Walter Newlyn and Bill Phillips met as students at the London School of Economics in the late 1940s after both having wartime service experiences.
Phillips wrote a paper exploring stock and flow relationships in economics using hydraulic system analogies. Newlyn recognized the potential in turning this into a physical machine model.
During the 1949 Easter vacation, Newlyn worked with Phillips to design an actual hydraulic machine to model the economy. Newlyn contributed his economic knowledge and Phillips contributed his background in engineering.
They built a prototype model together over the summer of 1949. It was demonstrated at LSE later that year and then taken to the University of Leeds where Newlyn worked.
This original prototype became known as the Newlyn-Phillips Machine and was the first physical model of the economy created using hydraulic systems to represent monetary flows and stock-flow relationships. It helped launch Phillips’ career in applying control theory to economics.

Here is a summary of the key points about the academic papers Phillips and Newlyn wrote on the launch of the hydraulic Machine model:

Phillips’ 1950 paper “Mechanical Models in Economic Dynamics” focused on the engineering aspects of the hydraulic Machine model. It provided strong technical details on how the model was constructed.
Newlyn’s 1950 paper “The Phillips/Newlyn Hydraulic Model” concentrated more on the economics and monetary circulation aspects modeled in the Machine. It looked particularly at how the Machine modeled money flows in the economy.
The papers showed how Phillips and Newlyn brought complementary skills and knowledge together in developing the Machine model. Phillips contributed the engineering expertise while Newlyn focused more on the economic theory.
The idea for the model originated from Phillips, who first conceived of modeling supply/demand interactions hydraulically in an undergraduate paper in early 1949. This caught Newlyn’s attention and sparked their collaboration.
Phillips’ 1949 paper diagramed economic concepts like supply/demand using hydraulic analogies, inspired by a similar plumbing diagram from Boulding. This provided the basis for developing the actual Machine model.
So while both men contributed, the original concept and imaginative leap to an analogical hydraulic model came from Phillips, which Newlyn then helped refine and translate into a physical working model through their collaboration. The papers highlighted their joint contribution but innovative origins.
Phillips created an initial diagram representing monetary flows in an economy analogous to a hydraulic system. This captured Newlyn’s interest as it implicitly incorporated the important element of time lags between inflows and outflows.
Newlyn saw potential to turn the diagram into a machine that could simulate the economy over time, introducing a third dimension of time.
During the 1949 Easter vacation, Phillips and Newlyn worked together on specifications for such a machine. Newlyn drew up the design for a “full economy version” incorporating all major monetary flows, sectors, and stocks. This went beyond Phillips’ original diagram.
The design involved mechanizing the flows and incorporating physical elements like floats and levers that could represent and potentially manipulate variables like savings, investment, interest rates over time.
This process of designing the machine captured how the economic relations unfold through time, which informed Newlyn’s later theoretical work on monetary macroeconomics and the role of time lags. It represented an imaginative leap from a static diagram into a dynamic, mechanized simulation of the monetary system.
The blueprint was drawn by Walter Newlyn on May 5, 1949 probably to secure funding from his department head at Leeds for building the Machine.
It shows the basic components and structure of the Machine, though some rearrangement occurred later.
Newlyn’s blueprint design represents a fully realized economic model, incorporating key economic features like:
- Circular flow of income/expenditure
- Other economic sectors (overseas, government)
- Separating flows (transactions) from stocks (money holdings)
It indicates how macroeconomic relationships would be incorporated via “slides” to control flows.
Stocks of money, foreign exchange rates, interest rates would be measured on calibrated scales.
It showed the hydraulics and controls but was not a detailed engineering drawing.
Newlyn privileged the economics of the model while Phillips was more involved in engineering aspects.
Their design used contemporary macroeconomic terms but defined some elements differently than standard definitions to fit the hydraulic medium.
The design allowed demonstration of different theoretical perspectives like Keynesian, Wicksellian, and Kaleckian views.
Newlyn’s expertise in macroeconomics and monetary economics was key to transforming Phillips’ initial diagram into a fully realized economic model.

Here is a summary of the key points in the passage:

In the summer of 1949, Bill Phillips and Walter Newlyn built the first prototype of their hydraulic model of the economy (known as “the Machine”) in the garage of Bill’s friends in Croydon, South London.
Newlyn’s role was primarily as a “craftsman’s mate”, helping with tasks like sanding and gluing perspex pieces. But they also took time to discuss the economic theory behind the model.
Both Phillips and Newlyn brought useful skills to the project. Phillips had experience inventing things and fixing mechanical/electrical problems. Newlyn also had technical skills from his wartime training and experience keeping vehicles running in difficult conditions in Africa.
They creatively used spare parts from things like bomber planes to build components of the Machine, like using perspex from plane windows and a windshield wiper motor.
Once built, maintaining the Machine and getting it to operate properly required continued technical skills from Phillips and Newlyn.
They worked to calibrate the Machine so the time it took to reach equilibrium after a “policy shock” correctly modeled economic time lags, though this proved difficult.
The passage emphasizes the truly collaborative nature of their work and how their complementary skills and knowledge allowed them to successfully design and build the first prototype of their innovative hydraulic economic model.
Irving Fisher created both arithmetic and analogical models of his monetary circulation equation in his 1911 book. This included a mechanical balance model.
The mechanical balance model served as an analogical representation of the arithmetic model, showing how money balances with goods exchanged on either side of the balance, similar to his monetary circulation equation.
Key positive features of the analogy included money balancing with goods exchanged, similar to the equation. Exploring neutral features, like how to map velocity of circulation and prices, required creativity to fit the economics onto the mechanics.
This allowed Fisher to gain important new insights, like investigating changes at the aggregate level. While not a real machine like the Phillips model, it conceptually demonstrated how the system works and relationships within it.
The Phillips model similarly used an analogical representation through their hydraulic machine to gain new understandings of monetary flows and the macroeconomic system in a substantive, rather than just mathematical, way. Both Fisher and Phillips/Newlyn leveraged analogical models to develop new insights.
Fisher developed the concept of a “weighted average” to map aggregate prices and quantities onto his equation of exchange. This led to seminal contributions to index number theory.
Mapping economics onto the mechanical diagram of a balance also led Fisher to use the model for other measuring purposes and reasoning about monetary economics debates.
Working with the balance model prompted two-way reflection - translating economics into the model, and then insights from reflecting back on features of a real balance.
Reflecting back, Fisher noticed balances don’t necessarily maintain equality if one side changes, unlike his accounting identity. He also drew on oscillations to integrate cycle theory.
This two-way reflection through both sides of the analogy led Fisher to reinterpret his equation as a tendency to equilibrium rather than continuous equilibrium, and incorporate cycle theory.
The Newlyn-Phillips Machine similarly reconciled theories through translating economics into hydraulics in the prototype, demonstrating that interest rate determination involves both liquidity preference and loanable funds working together.
Phillips and Newlyn created an economic machine model called the Phillips Machine to help understand and represent macroeconomic dynamics and interactions between variables like income, consumption, savings, investment, etc.
The machine modeled the economy using a hydraulic system where water represented money flowing through the system. This allowed stocks and flows to be represented together dynamically over time, addressing limitations of existing diagrams and models.
Using the machine, economists could experiment by manipulating variables and observe the effects on other parts of the system. This helped deepen understanding of issues like time lags between variables.
Later economists like Vines who studied the machine agreed it provided unique insights compared to static theoretical models. It stimulated new thoughts by making the dynamics and interactions more visibly represented.
Insights from working with the machine seemed to influence Phillips and Newlyn’s later economic works, as they brought over ideas about circulation, dynamics, and time from how the hydraulic model behaved to their economic theories. So the machine had a creative and cognitive role beyond just representing the economy.
The Newlyn-Phillips hydraulic machine model of the economy was a highly inventive and original modeling approach that provided new insights for both economics and the engineers who built it.
Reflecting between the hydraulic machine analogy and economics allowed for fruitful ideas to develop in both directions. The engineers gained understanding of macroeconomic dynamics while economists developed insights from the engineering perspective.
Though the specifics of the machine are now forgotten, it introduced concepts that became standard parts of economics. However, the machine itself was difficult to work with and depended on its inventors.
The machine captured the public imagination even for those who never saw it. Its depiction in cartoons conveyed its personality as an eccentric yet lively economic system operated by scientist attendants.
Economists enjoyed its boldness and novelty, though few saw it work in person. It remains one of the only economic models to permeate popular culture through descriptions in media over time.

So in summary, the Newlyn-Phillips hydraulic machine was a highly creative modeling approach that provided new insights for both its engineer-builders and economists, though its actual workings are now largely forgotten despite enduring in the public imagination. The bidirectional reflection between its analogy and economics yielded fruitful ideas.

Here is a summary of chapter 8, page 10:

This section discusses how Ragnar Frisch developed one of the early mathematical models of business cycles in the 1920s-1930s. At the time, accurately modeling business cycles was a theoretical and practical challenge.

Frisch wanted to create a model that could endogenously generate cyclical patterns in overall economic activity. He began with a visual “tableau économique” depicting the elements and flows of the economy. He then developed a simpler mathematical “macro-dynamic system” with the key variables and relationships.

This systems had two important properties. First, it could produce cycles in economic activity internally, through the interaction of elements in the model over time. Second, any cycles generated would gradually die out on their own over time. These two features were important to Frisch because real-world cycles existed but classical economists believed the economy would reach equilibrium if left undisturbed.

The model therefore had the necessary theoretical and mathematical “resources” to function like a machine that could endogenously create and then dissipate business cycles through its internal dynamics. In developing this initial macroeconomic model, Frisch employed both visual diagrams and mathematical formalization to shape the representational abilities of the model.

Ragnar Frisch created a simple mathematical model of the business cycle that could produce cycles matching the lengths of real economic cycles. However, the cycles in his model were too neatly cyclical compared to the irregular patterns seen in real data.
This led him to ask how his deterministic model could be modified to explain irregular fluctuations in the real world. He was inspired by stories and statistical experiments showing that random shocks could disturb harmonic processes and produce irregular cycle patterns similar to economic data.
Frisch then augmented his model by introducing random shocks, allowing it to propagate these shocks over time in a way that produced simulations matching the jagged patterns in real data.
He drew on analogies like Wicksell’s rocking horse being randomly struck to shape how random impulses were incorporated into the model.
Schumpeter’s theory of innovation-driven cycles provided Frisch with an economic interpretation for how cycles were maintained in the real economy.
Younger economists like Meade and Samuelson also used simple algebraic/geometric models to try to understand and represent Keynes’ General Theory, which combined words and mathematics opaquely. Their models aimed to capture the essence of Keynes’ ideas and allow comparison to classical theory.

(i) The conditions necessary for equilibrium:

Meade outlines 7 assumptions about specific elements in the economy (e.g. wages are the prime cost)
He then lists 8 conditions under which the economy based on those assumptions will be in short-period equilibrium (e.g. prices equal marginal costs, total income equals wages plus profits)

(ii) The conditions necessary for stability of equilibrium:

Meade constructs a mathematical model based on the 8 relationships
He checks that the model world will return to equilibrium following a change and that the equilibrium point is stable

(iii) The effect on employment of changes in certain variables:

Meade analyzes 4 cases looking at the impact on employment of changes in interest rates, money supply, wages, and savings rates
He traces the impact of each change through the model’s relationships to determine the effect on demand for labor, which was a key issue given the unemployment of the time

The key aspects are Meade establishing the equilibrium conditions, checking the model’s stability, then using the model to trace the impacts of certain changes on employment by following the causal links between variables in the model.

Here is a summary of the key points about ‘amic”:

Amic’ is a Latin word meaning friend. It is the root of words like ‘amicable’, ‘friendship’, and ‘amity’.
It refers to friendly relationships, friendship, being friendly or amiable towards others.
It implies positive feelings of goodwill, kindness, caring and benevolence between individuals or groups.
In its original Latin context, it referred specifically to people who knew and cared for each other, rather than just acquaintances. There was an implicit sense of closeness, trust and loyalty.
Derivatives of amic’ today still convey a sense of warm, caring interpersonal relations characterized by mutual support, understanding and benevolence rather than conflict or indifference.

So in summary, amic’ and its derivatives describe friendly, supportive relationships between individuals or groups based on positive feelings of goodwill, caring and trust. It implies a degree of closeness beyond a casual acquaintance.

Here is a summary of the modelling practices of Jan Tinbergen and the younger economists of Wicksell’s Stockholm school:

Jan Tinbergen was a Dutch economist known for his pioneering work in econometric modelling and policy analysis in the 1930s-1950s. He is considered one of the founders of macroeconometrics and helped develop techniques for estimating economic relationships and policy impact.
The younger economists at Wicksell’s Stockholm school in the early 20th century focused on developing dynamic theoretical models to analyze economic fluctuations and the effects of monetary and fiscal policy. They were influential in advancing Keynesian-style macroeconomic modelling.
Key figures included Bertil Ohlin, Dag Hammarskjöld and Gunnar Myrdal. They built structural macroeconomic models to study how different sectors of the economy interact over time in response to policy changes.
The Stockholm school models incorporated multiplier-accelerator relationships between investment, output and consumption similar to those explored later by Samuelson. They used the models to simulate the dynamic effects of policy interventions.
Tinbergen and the Stockholm school were pioneers in developing comprehensive macroeconomic models and using both theoretical/analytical and quantitative/simulation approaches to policy analysis. Their work helped lay the foundations for modern macroeconometric modelling and policy evaluation.
The internal dynamic of a model refers to the deductive logic, mathematical rules, and relationships within the model itself. A model has an internal logic that allows it to demonstrate answers to questions.
However, a model needs an external dynamic in the form of questions posed by scientists to make use of its internal dynamic and provide demonstrations. Models do not solve or manipulate themselves without being prompted by external questions.
Hicks developed his IS-LM model in an attempt to compare Keynes’ theory to classical economic theory and identify what was novel about Keynes’ approach. He started with verbal descriptions and symbolic equations but found these limited.
By developing diagrams showing the relationships between income, interest rates, savings, and investment, Hicks was able to gain new conceptual insights. This opened up new analytical dimensions that allowed him to better understand and define Keynes’ innovations.
The IS and LM curves in his diagrams provided a way to represent different economic states and compare theories based on where the curves intersected. This diagrammatic form of modeling had lasting influence and helped establish Keynesian economics.

So in summary, the internal dynamic is the model’s deductive logic and rules, while the external dynamic comes from the questions posed by scientists, which prompts the model to demonstrate answers and insights. Hicks’ IS-LM model is an example of how developing new dimensions through modeling can generate new conceptual understandings.

Hicks created a diagrammatic model (IS-LM curve) to help elucidate the differences between Keynes’ theory and classical economics. He saw the model as a “physical apparatus” that could demonstrate relationships and answer questions.
The model gave Hicks resources to go beyond some of Keynes’ simplifications. It allowed him to make the investment equation dependent on both interest rates and income, creating a more gradual rising LL curve.
Hicks used the model to represent different theories from other economists and different states of the economy during the Great Depression. It enabled him to explain, demonstrate, and tell stories about the theories and economic conditions.
While flexible, Hicks noted the model was still a “skeleton apparatus” that couldn’t capture everything. However, its ability to represent multiple theories and scenarios through demonstrations is what gave the IS-LM model, later renamed by Hansen, such longevity and influence. It provided a useful method for economic inquiry.

The passage discusses the role of narratives or stories in economics modeling work. It argues that while economists usually recognize mathematical reasoning in their models, they also commonly use narratives when explaining or investigating models. Narratives take the form of extended examples that simplify real-world situations and map them onto the mathematical representations.

Telling model stories is not just a rhetorical practice but an epistemic one. Narratives help economists understand the relationships and logic within the world of the model. They provide a way to interpret model demonstrations and results in economic terms and link the model to the real world. This helps economists capture the “heart of the matter” by exploring both the model world and the represented real world.

Specifically, model construction can be seen as “fitting out” underlying economic theories or hypotheses in concrete form, just as fables fit out abstract morals. Using models to generate narratives then allows economists to grasp particular implications and understand the general theory being represented. The narratives make the abstract relationships within models graspable by providing concrete examples and stories. This process of using models reveals insights that model builders may not have anticipated in advance.

Economists use narratives/stories to understand their models and communicate the outcomes of complicated models more easily. Stories are how economists comprehend what is captured in their models of the economic world.
When developing models, economists represent real-world situations theoretically. But models alone don’t fully explain how the world works. Narratives enable economists to understand their theories and form connections from models back to the real world.
Model narratives provide possible links between model demonstrations and real-world events/processes/behavior represented by the model. They show how models can apply to the world and offer insights/understandings of how the world works.
While early models like Hicks’, Samuelson’s, and Meade’s explored theoretical worlds, their stories also spoke to real economic problems of their time like the Great Depression. Model narratives were not just about theory but about applying Keynesian ideas to the real world.
Model narratives describe economic situations at a general level but stories told using models can describe very specific, detailed, differentiated particular events. Narratives link the general to the particular and particular to the concrete realities.
Narratives associated with economic models can function as an informal test of the validity and explanatory power of the models.
Good model narratives demonstrate consistency (facts don’t contradict each other) and coherence (events are logically connected). This allows narratives to pull disparate elements together and provide accounts of the real world.
Economists judge models based on the quality of the narratives they can generate. Meaningful narratives tell stories that are theoretically meaningful and reveal interesting economic behavior in the model.
Plausible narratives map adequately to characteristics of real-world phenomena the models aim to describe. Economists connect models to specifics of the real world to test narrative plausibility.
Generating and critiquing model narratives allows economists to both explore the implications of their models and create coherent accounts of economic events and patterns in the real world. Narratives act as a testing bed for how well models represent reality.
Economists first need to ensure their small mathematical models are plausible and ask meaningful questions before using them to tell stories about real-world events.
For a model to be plausible, it needs to provide a loose fit to the world it represents and offer some insight into why the economic situation is a certain way.
Frisch’s model of business cycles was seen as plausible because it could produce cycles that fit real-world data patterns and explained cycles through innovations rather than just mechanics.
Economists don’t take narrative plausibility as proof of a model’s truth, but storytelling does involve inference about how models apply to the real world.
As time passes, what counts as meaningful and plausible in a model can change based on new economic theories, evidence, and knowledge. Factors like different events needing explanation also impact judgements of plausibility.
Models allow economists to offer explanations somewhere between general laws and specific cases, balancing abstract and concrete levels of analysis.
Economic models are more flexible in content than laws, so they are regularly adapted to fit new problems or phenomena. New theories and concepts also prompt changes in models.
What counts as plausible is shaped by the “epistemic genre” of modelling that economists use. Different fields adopt different modes of reasoning (e.g. modelling, experimentation) that influence what seems reasonable.
Modelling forms the broader context in which economic models are judged as plausible. The effectiveness of model narratives in “explaining” the world depends on implicit and explicit scientific knowledge that is conceptual, empirical, historical, and theoretical.
Models are “tested” against this accepted knowledge within scientific community norms. They are found meaningful, plausible and credible, or lacking. This process is not incompatible with the idea that models are “fables” regarding laws but “parables” regarding the world.
The meaning and interpretation of a model for the “target system” (real world) has to be drawn out with the help of other information. But narratives are also important as the vehicle for interpretation and bring their own criteria like coherence and plausibility.

Here is a summary of the key points from the chapter:

The chapter explores treating model-based reasoning in economics as a form of experimentation. It looks at how economists can experiment both within model worlds and through using models in laboratory or empirical experiments.
When experimenting within models, economists manipulate the assumptions and parameters of the model to derive answers to questions about that model world. They explore the implications of different scenarios.
Models also play roles in laboratory and field experiments, where they may structure the design, variables being manipulated or observed, or set constraints. So models can be both the object of experimentation and part of experimental designs.
By experimenting within models, economists aim to demonstrate answers to questions of interest. They then need to make inferences from the model world to the real world. The chapter discusses the challenges of this “inference gap.”
Experimenting within models also prompts the development of more generic conceptual categories for analysis that define and limit the domains for model-based inferences.
The supply and demand model is used as a key example to illustrate how economists experiment within models and use models in experiments regarding market mechanisms.

So in summary, the chapter frames model-based reasoning as a form of experimental inquiry and explores the different ways experiments can involve or utilize models, as well as the challenges of moving from model results to claims about the real world.

The chapter examines the development and use of supply and demand models through experiments conducted with the models. It uses the supply and demand model as an exemplar to show how modeling became standardized in economics.
Early pioneers of the supply and demand diagram included Antoine-Augustin Cournot in 1838 and Karl Heinrich Rau in 1841, but its use remained uncommon until the late 19th century.
Hans von Mangoldt in 1863 and Fleeming Jenkin used supply and demand diagrams in their work, showing an early evolution of using models and diagrams to construct arguments. Mangoldt in particular used numerous diagrams to demonstrate his arguments.
While Mangoldt’s initial use of diagrams did not involve experiments, some of his diagrams like figure 8 showed how he conducted experiments by manipulating the curves to examine different effects.
The chapter examines how model experiments evolved from early uses like Mangoldt’s to become a standard way of reasoning and understanding economic concepts and interactions through manipulating models.
The passage discusses how the natural price of goods can shift upwards or downwards depending on two opposing tendencies of progress: y, which tends to push up natural price, and the former (unspecified), which works in the opposite direction.
Whether natural price goes up or down depends on which of these two tendencies is more prevalent.
Mangoldt’s use of diagrams is highlighted as providing clear demonstrations of how differing assumptions or changes can lead to different outcomes, allowing him to explain why there is no simple general answer to how prices are affected by progress.
The text then discusses how Mangoldt used more complex diagrams, equations and numerical experiments to analyze interdependent supply and demand for two goods, which helped elucidate cases that were difficult to follow through verbal reasoning alone.
Mangoldt’s innovative use of model-based reasoning is praised, though it was unfortunately cut from later reprints, hindering understanding of his work. The passage examines how he used models experimentally to answer questions.
The text discusses Fleeming Jenkin’s use of supply and demand diagrams in 1887 to illustrate the theoretical laws of supply and demand using the example of the corn market in Britain.
Jenkin presented imaginary supply and demand curves for corn, showing quantities on the vertical axis and prices on the horizontal axis. This was contrary to later conventions but consistent with mathematical practices of the time.
The curves used actual quantities (quarters) and prices (shillings) to make the illustration empirically relevant to the debates around corn prices and tariffs in Britain.
The text then discusses Alfred Marshall’s use of supply and demand diagrams in his 1890 work Principles of Economics. Marshall standardized the positioning of prices on the vertical axis and quantities on the horizontal axis.
Marshall posed four questions and conducted ten “model experiments” using diagrams to demonstrate points, and six associated “mental experiments” where he pointed to answers without diagramming.
The experiments allowed Marshall to analyze the impacts of changes in demand, supply facilities, taxes/bounties and discuss wider theoretical and policy implications. The diagrams enabled complex reasoning and classifications.
Marshall’s use of multiple experiments with the model demonstrated his facility in experimentally manipulating and reasoning with the supply/demand diagrammatic model.
Marshall, Mangoldt, and Jenkin used supply and demand diagrams in their economic analysis as a way to manipulate models and reason through problems. They used these diagrammatic models to answer “what if” type questions.
While they used similar basic supply and demand diagrams, their conceptual frameworks and understandings of demand, supply, and their intersections were not entirely the same.
By asking questions and manipulating the models through “model experiments”, they were able to classify different market situations, develop generic categories of cases, and further conceptualize the laws of demand and supply as they applied in different circumstances.
This model experimentation allowed them to explore, analyze, define, and divide the applications of supply and demand in a way that generated new conceptual elements and categories, developing economic theories about markets.
Later, starting in the mid-20th century, economists began conducting classroom experiments using actual people and markets to test assumptions that had previously been explored through diagrammatic model experiments alone. This brought models into the laboratory setting to empirically examine whether people actually behaved as assumed in the theoretical models.

Here are the key differences between rice in real world markets vs idealized economics models/theory:

In the real world, rice markets have many complicating factors that economics models abstract away from, such as imperfect information, transaction costs, behavioral factors like biases and heuristics, political influences, and more. The real world cannot be perfectly controlled or isolated like in economic models.
Economics models and theory rely on “abstract models” and idealized assumptions like perfect competition, rational actors, instantaneous equilibrium. They aim to isolate the effects of particular variables by holding others constant.
However, in reality markets have “steps” and gaps rather than smooth continuous supply/demand curves assumed by models. Reserve prices are discrete numbers rather than smooth curves.
Experiments by Chamberlin and Smith found real world transactions differed from equilibrium predictions - average prices were lower and quantities higher than model predictions. This challenged the idea that markets naturally tend towards equilibrium.
Real actors have imperfect knowledge and reactions, unlike the omniscience assumed in models. They may not react or converge as smoothly or quickly to equilibrium as theorized.
The real world rice market incorporates many behavioral, social, political and historical factors not captured by abstract theoretical models. It is a complex system not fully explained by isolated theoretical variables.

In summary, economics models provide idealized conceptual tools but fail to fully capture the complexity of factors influencing real world rice markets behavior. Experiments highlighted differences between theory and real market outcomes.

Here is a summary of Collins’ investigation into tacit knowledge in creative laboratory work as presented in Collins (1990):

Collins studied crystal growing laboratories and observed that skilled crystal growers developed tacit knowledge about the subtle influences on crystal growth that could not be fully articulated. Even experienced growers had difficulties fully describing in words what influenced their judgements.
Crystal growing is a complex process influenced by many small changes in conditions like temperature, vibration, air currents, etc. It requires close monitoring and making ongoing adjustments to manipulate these subtle influences to produce high quality crystals.
Through extensive hands-on experience, growers developed a feel for how the apparatus and crystal reacted that allowed them to make the right adjustments intuitively without always being able to explain their reasoning. Their judgements were based on patterns they recognized but could not necessarily describe.
This tacit knowledge developed through experience was a key part of the skill of expert growers. It allowed them to produce better results than someone just following written instructions, and formed an important part of the expertise involved in creative laboratory work like crystal growing.

Here is a summary of the key points about model experiments and laboratory experiments:

In a laboratory experiment, the scientist creates a controlled real world within an artificial environment. They place controls on the experiment through the experimental design and setup.
In a model experiment, the modeler creates an artificial world within the model. They place controls through how the model is designed and the assumptions made, like ceteris paribus conditions.
Laboratory experiments demonstrate results experimentally, through intervention and observation in the controlled real world environment. Model experiments demonstrate results deductively, through logical manipulation of the model.
Inferences can be made more easily from laboratory experiments to similar real world situations because the experiments use real human and material components, just in a controlled setting. However, the narrow scope of controls limits inferences to exactly matching situations.
Inferences from model experiments to the real world represented in the model have a wider potential scope but are weaker in validity. This is because the model uses artificial representations rather than real world components, so how well it matches the real world is uncertain.
The narrow controls in laboratory experiments reduce confounding factors, while the wider scope but looser controls in model experiments allow for more potential surprises in results.

In summary, laboratory experiments have stronger inference validity but narrower scope, while model experiments have wider potential scope but weaker inference validity due to the artificial nature of the model world. The type of demonstration and materials used impacts the inference that can be drawn to the real world.

Economists like Chamberlin and Smith were able to make inferences about behavior under certain market rules based on experiments, but the validity was limited to similar situations.
Mathematical modelers can potentially make broader inferences because their models are created through abstraction and simplification, not observed realities. This allows more experimental variation and control.
However, the abstractions and simplifications also limit applicability to specific real-world cases.
Models operate at a generic level of classes/types, so experiments have some degree of both particularity and generality. Inferences can apply to real situations sharing common traits but lack details for any one case.
Successful model experiments depend on accurately representing elements relevant to the question being asked. But economists don’t know if their models are accurate representations of reality.
Both laboratory experiments and model experiments rely on controls and artificial environments, raising challenges for inferring real-world applicability. Formal inference procedures are lacking.

So in summary, models allow more generalization than experiments but lack validity for specific cases, and establishing accurate representations or inference criteria remains problematic. Narratives play a role in establishing credibility of inferences.

The passage discusses “hybrid experiments and simulations” that use models in experiments where real-world experiments and model experiments alone are problematic.
It focuses on the work of economists Cars Hommes and Joep Sonnemans, who wanted to study complex market behavior beyond lab experiments or mathematical models alone.
In their “virtually experiments,” human participants predicted future prices in a hypothetical market from their computers. These predictions were inputs into an underlying mathematical market model specifying supply and demand.
The model then used the predictions to calculate “realized” market prices, which were fed back to participants. Parameters in the model were varied experimentally.
This hybrid approach allowed investigating market dynamics at a complex level, taking advantage of both human inputs and a mathematical model structure, while avoiding issues with modeling or lab experiments alone. The model served as both a framework linking participants and the calculation engine during the experiment.
The experiments conducted by Cars Hommes and Joep Sonnemans used a laboratory experiment format to explore learning and behavior in a mathematical market model.
Participants had to predict prices in the model market, facing random noise and changing demand/supply conditions between experimental runs. This created variations that participants had to respond to.
While artificial in some respects due to rules and structured responses, the experiments also had real-world aspects from the natural variation introduced by human participants and their predicaments.
The mathematical market model used was based on established “cobweb” models from previous empirical work. However, parameters were chosen to allow a range of outcomes based on participant responses.
Participant strategies varied widely, with no single dominant strategy, limiting the inferences that could be drawn. Combined in the model, strategies could still only be classified into 5 groups.
These experiments sit at the boundary between laboratory and model experiments by introducing real human responses into a mathematical market model. This allowed comparison to “traditional” model experiments using theoretical agent behavior.
Later experiments by Hommes and Brock used mathematical decision rules to simulate traders following different approaches like fundamentals vs. technical analysis. This pushed these experiments closer to computer simulations.
The passage discusses different types of model experiments and hybrid experiments that mix real-world and abstract elements.
In some hybrid experiments, there are no real people and it is close to a pure model experiment. In others, like Hommes and Sonnemans’ work, both laboratory and model elements are incorporated at the design and experimental stages.
Models can play many roles in experiments - sometimes as the object of experimentation, sometimes as part of the experimental apparatus, and sometimes creating the economic world they represent.
There is a temptation to view model experiments as predetermined due to the model building, but economists can still be surprised by results from varying model parameters. Surprise indicates new learning.
In model experiments, surprise comes from ignorance about the model world. In laboratory experiments, ignorance is about the real world behaviors and subjects, allowing for possible confoundment or unexpected results beyond the model.
Maintaining real-world inputs in experiments allows for more empirically rich, unexpected, and genuinely learning outcomes through potential confoundment.
Economic experiments aim to give participants some freedom to behave as they choose within the experimental setup, rather than strictly controlling their behavior to match economic models. Strict control risks making the experiment more about validating the model than learning about real-world behavior.
Model experiments, which test theories in abstract simulated worlds, allow less inference about real-world behavior than laboratory experiments that use real-world materials. However, model experiments can still surprise researchers by producing unexpected results, leading to new insights and refinements of economic concepts and theories.
The possibility of surprise and confounding results is an important epistemological strength of laboratory experiments compared to model experiments. Laboratory experiments have potential to uncover new phenomena not predicted by existing theories by observing real-world behavior.
Both model and laboratory experiments play a role in developing economic understanding. Model experiments often lead to new theorizing within the model, while laboratory experiments using real materials have stronger inference for the real world. A balance is needed between control and allowing freedom of participant behavior.

Here is a summary of the provided text:

Simulation emerged as a new technique in social sciences in the 1960s, encompassing a broad range of practices like role-playing experiments, use of computers/models, statistical data, etc. Though diverse, these were all grouped under the umbrella term “simulation.”
The emergence of simulation combined older experimental traditions from statistics with newer experimental modes in social sciences as well as the new computing technology.
The historian analyzes two figures influential in developing simulation techniques in economics - Martin Shubik and Guy Orcutt. Their histories help understand how simulation constituted older and newer techniques/ideas into a varied set of simulation types.
Models play an important role in the technology of simulation. The text aims to understand how models fit into simulation and how simulation fits into the history of reasoning modes in economics.
The historical analysis focuses on the American context in the post-World War 2 period amid the rise of computing and Cold War research technologies. It examines how simulation emerged from various traditions to become a new method of modeling that combined these elements with experimentation.

This summary discusses simulation as a methodology that emerged in the social sciences in America around 1960. Some key points:

1960 is seen as a moment when different types of experiments with models (both real experiments and computer simulations) came together under the single term “simulation”.
An AER symposium in 1960 and a 1962 book brought attention to simulation as a new way of doing social science. Literature at this time felt the need to explain and promote simulation.
Shubik’s 1960 bibliography showed simulation spanning many disciplines, from management to political science to economics. It included things like war games, business games, and role-playing experiments.
The terms “simulation” and “gaming” were difficult to clearly separate, as simulation involved both simulated environments and processes. It was a broad research approach.
Guetzkow’s 1962 book also showed the cross-disciplinary nature of simulation, including topics like flight simulators, RAND role-playing experiments, and computer models from various fields.
There was a sense at this time that simulation provided a “microscope” to closely examine models in a way not possible with other analysis methods. It allowed new insights into models and their ability to mimic the real world.
Martin Shubik developed an interest in game theory after discovering Von Neumann and Morgenstern’s book while browsing the library at Princeton in 1949.
He was involved in the development of game theory as a student and early researcher at Princeton in the late 1940s/early 1950s, where some of the main research was happening.
By 1953 he had published a book of readings in game theory. While still developing his academic career in game theory, he was also involved in gaming experiments and simulations.
His 1960 bibliography on simulations and gaming showed the intersection between social sciences, defence establishments, and the growing military-industrial-science complex of the Cold War period. Many studies were classified or related to defense contractors.
Importantly for Shubik, “gaming” referred to experimental role-playing simulations with people, not game theory, which was a separate mathematical field. Game theory was absent from his bibliography despite its connections to institutions like RAND that promoted both.
Shubik was an influential early researcher and historian involved in developing both game theory and experimental simulations using games and gaming.

The text discusses Martin Shubik’s career and contributions to the field of economic simulation. Some key points:

Shubik had a background in mathematics, electronics, and served in the Royal Canadian Navy. He was educated in England and Canada.
He helped develop game theory and its application to industrial economics and the behavior of firms. His 1959 book Strategy and Market Structure was influential in integrating game theory into economics.
He recognized the value of experiments and gaming to test theories from game theory. In the late 1950s he started collaborating on early experiments with Siegel and Fouraker.
In the 1960s he conducted his own experimental work and went to IBM to develop business games that could be both training tools and experimentally analyzed.
His 1960 publications helped survey the emerging field of economic simulation, which integrated his knowledge from game theory, experiments, and business gaming.
Models are often hidden aspects that economists use to represent resources and behaviors in simulations. Shubik helped advance the use of models, simulated environments, and studying simulated behavior in economics.

The book discusses military simulations and games (known as MSGs) that were used by the U.S. military. While the book was copyrighted by RAND, a military-focused think tank, it was published by Harvard University Press, an academic publisher.

The book provides examples of different types of MSGs used in the 1950s-1960s time period. Some involved detailed modeling of real industries or firms based on empirical data, while others used entirely hypothetical or fictional models. Computer modeling and simulations played an important role, especially in “man-machine” simulations that combined human decision-making with complicated mathematical models.

Business games that placed human participants in decision-making roles to simulate industries were also discussed. These provided insights into topics like team actions, decision-making behavior, and allowed the study of oligopoly industries. Overall, the book examines how models, empirical data, computers and human participation were combined in various ways within military simulations and games during this early period.

Guy Orcutt pioneered the use of microsimulation in economics in the 1960s as a new research method. His microsimulation approach uses simulation as an essential part of the research method, unlike other uses of simulation in economics at the time which were more complementary.
Orcutt was originally interested in engineering and physics but switched to economics. He built one of the first electronic computers to do statistical analysis. This experience informed his later work.
Orcutt questioned the validity of using aggregate time series data and macroeconomic models for policy analysis and simulation given problems with correlation in the data.
His microsimulation method created a “sample” of virtual individuals using census and survey data to represent the US population. It then simulated demographic changes like births, deaths, marriages over time at the individual level.
The virtual individuals’ characteristics were based on empirical evidence, while demographic behavior probabilities came from statistical evidence. Monte Carlo techniques randomly determined which events occurred for each individual based on the probabilities.
This combined elements of Tinbergen’s model-based approach and Slutsky’s use of statistical randomization techniques to generate artificial time series for analysis. It allowed Orcutt to avoid problems with aggregate data correlations.
Orcutt’s microsimulation models incorporated individual decision units from demographic data combined with their labour force participation, spending and saving behaviors.
He generated random selections by computer rather than manual drawings or tables of random numbers, representing an early adoption of Monte Carlo techniques using computers.
The model was designed with “block recursiveness” in mind, acknowledging that decisions are made sequentially over time as advocated by Wold and Simon. Individual family units made decisions in monthly “blocks”.
The model had “pluggable components” allowing parts to be tested and revised independently, reflecting Orcutt’s background in electrical engineering.
It used a probability approach developed by Haavelmo to treat the sample as a probability sample representative of a larger conceptual population.
Factors like Orcutt’s statistical training, experimentation with computers, experiences in econometrics debates, and engineering background all influenced the design of his pioneering microsimulation models and the new “microscope” they provided for economics.

Here are the key points about how economics uses digital computers and simulation according to the passage:

Digital computers allow economists to study large amounts of data at various levels of aggregation. This provides a “viewing equipment” analogous to a microscope.
Computers enable the use of more complex, “realistic” models that do not have to be solved analytically but can be analyzed numerically.
Computers enable simulation, which is the real focus rather than the computer itself. Simulation provides an exploratory, investigative tool to make economic phenomena observable in a way analogous to a microscope.
Early on, simulations were used for statistical studies of whole aggregated economies using macroeconometric models. Computers allowed processing much larger datasets and calculations.
However, aggregation was seen as problematic. Simulation’s computing power in the 1960s allowed economists to move away from aggregates to smaller individual units of analysis.
Simulation is understood as extending observation beyond just scale to reveal hidden details, structures, and aspects of economic phenomena in an “active” or “invasive” way similar to other scientific instruments like microscopes.

So in summary, the passage discusses how digital computers and the technology of simulation provided economists new tools for observation, experimentation, and investigation analogous to microscopes, allowing them to analyze more complex and disaggregated models and simulations of economic phenomena.

Simulation with computers allows economists to analyze the economy at a finer-grained level than previously possible, examining specific industries or individual behaviors over short time periods rather than only large aggregates.
Computer simulation provides a way to “see” the economy both at the detailed level through micro-level models, as well as aggregate back up to the big picture by combining individual units of analysis.
Studying individual cases does not reveal the interactions and interrelationships between units, which computer simulation of models can uncover from examining interacting units.
The “specimens” being examined through simulation are not natural economic entities themselves, but models representing things like families, firms, industries.
Building accurate models for simulation requires extensive data collection and preparation work to represent the real-world behaviors and relationships being modeled at a fine-grained level of detail, similar to preparing biological specimens for microscopic examination.
Computer simulation thus allows economists insights into aspects of the economy previously hidden at higher levels of aggregation by dynamically interacting micro-level models representing individual units.

Here is a summary of the key points from the excerpt:

Clarkson and Simon argued that computer simulations allowed for a more direct match to individual decision-making processes, without needing to first develop a complex mathematical model. They could directly program symbolic decision-making logic.
Their simulation required fewer translations or transformations between the source economic concepts/data and the final simulation model, compared to other simulations described by Shubik that involved mathematical modeling.
The simulation was able to closely match how computers can process symbols just as human decision-makers process information.
While there is a tradeoff between preparing the simulation model and letting the computer do more work, the relationship between model preparation and the computer’s role is difficult to pin down precisely.
When used as “microscopes,” simulations harness natural laws through the programming/code to actively probe and reveal behaviors of the simulated economic system/model in a way that is analogous to how microscopes reveal structures by probing specimens.
Orcutt’s microsimulation model treated individual households/people as the basic components that could change statuses each period based on probabilities and behaviors programmed into the model. The simulation would reveal how these probabilistic individual behaviors played out at a population level over time.

Here are the key points about model experiment and model participation based on the passage:

In model simulations, scientists actively create variability in the model to observe its hidden characteristics and behaviors, unlike econometricians who rely on natural variability in real-world data. Scientists induce models to exhibit variability through organized experimentation.
It can be difficult to distinguish where the model as the “specimen” ends and the experimental laws/methods begin, as they are intertwined. Both define the “instrument” of observation.
Simulation packages used come as complex black boxes. Economists don’t need full understanding of how they work but can still use them to conduct experiments. This raises issues of distinguishing real model behaviors from artificial results of experimental manipulation.
The goal of model experimentation is often to gain insights about the real world from observing the model world. Inferences must be made from model behaviors to the target system it represents.
Slutsky’s random number simulation illustrates using experimental laws (arithmetic operations) to reveal hidden properties of the model (random numbers) and potentially make inferences about real-world phenomena like business cycles.

So in summary, model experimentation actively manipulates models to observe their properties, while aiming to ultimately learn about the real target system, but distinguishing real insights from artifacts is challenging.

Slutsky’s simulation showed that summing random numbers could generate patterns similar to real-world business cycle data, suggesting business cycles may be caused by random events.
Frisch interpreted this to mean the economy acts as an “adding device” that aggregates external random shocks to create business cycles seen in the data.
Inference from simulations involves not just mimicking output data, but examining the laws/model that generated the outputs.
Models need to mimic both real data and real-world laws/mechanisms to allow stronger inferences. Simply mimicking data is not enough.
Data on economic activity (B) may be recorded accurately, but are governed by bureaucratic/social rules of data collection, not economic laws directly.
Simulations can mimic real data through similar laws - e.g. if data are sampled randomly like Slutsky’s simulation. This suggests inferences should be about rules of data collection, not economic laws.
Analogizing simulations to microscopes clarifies that simulation laws need not match real-world laws, limiting inferences, but also shows how simulations investigate models using their own operative laws.
The passage discusses simulation models as a type of “microscope” for investigating and understanding the behavior of complex real-world systems. It analyzes how simulation techniques can provide insight into one domain by applying the laws and models from another domain.
Orcutt’s microsimulation model of the US economy in the 1950s is presented as an example. The model aimed to mimic key statistical features and behaviors of the real population and economy through probabilistic/statistical modeling techniques.
Two levels of “mimicking” gave Orcutt’s simulations credibility for making inferences back to the real world. First, the model specimens were designed statistically to represent features of the real population. Second, the governing laws/mechanisms of the model came from statistics and probability, which were empirically validated as descriptive of population behaviors even if not causally explanatory of individuals.
This close alignment between the representative model and the descriptive/statistical investigative laws allowed Orcutt’s simulations to credibly reveal behaviors of the virtual population that offered insights about the real population/economy it was designed to mimic. The independence of model and reality was compromised, but this enhanced the model’s “fruitfulness” as an analytical tool.

Here is a summary of the provided article:

The article is a literature review covering models and simulations in various disciplines like economics, science, and social sciences. It discusses several thinkers and their contributions to developing models and simulations as tools for research. Some key individuals and ideas mentioned include:

Ragnar Frisch’s work on dynamic economic models and simulations in the 1930s.
Peter Galison’s book on the role of images/models in microphysics research.
Studies by Gooding, Pinch and Schaffer on the role of experiments.
Greenberger et al.’s work on policy models and computer simulations.
Guala’s book on experimental economics methodology.
Early works editing simulation studies in the social sciences by Guetzkow.
Haavelmo’s probability approach to econometrics.
Hacking, Harré and others on the philosophy of instruments and experiments.
Development of modeling and simulation as processes across various disciplines.
Key texts in econometrics, economic modeling and others that advanced these fields.

The article provides an extensive literature review of the developments in various disciplines relating to models, simulations and their role in research over the 20th century. It covers work from economics, science, social science and philosophy of science perspectives.

Here are summaries of the key papers:

Samuelson Paul A. (1939) “Interactions between the Multiplier Analysis and the Principle of Acceleration”. Review of Economics and Statistics, 21, 75–8.

Discusses the interactions between the multiplier analysis (which analyzes how spending affects income and output) and the principle of acceleration (which describes how changes in income/spending growth rates affect one another). Finds that the two approaches are fully compatible and mutually supportive.

Schelling, Thomas C. (1978) Micromotives and Macrobehaviour. New York: Norton.

Explores how individual or “micromotive” behavior can result in aggregate or “macrobehavior” outcomes. Discusses self-organizing systems and how coordination or collusion can spontaneously emerge from uncoordinated individual interactions. Introduces several concepts/games used to study coordination and cooperation.

Shubik, Martin (1952) “Information, Theories of Competition, and the Theory of Games”. Journal of Political Economy 60:2, 145–50.

Argues that the explicit consideration of information is needed for theories of competition and game theory. Discusses different kinds of information structures and their implications. Calls for incorporating information issues into economic models of competition and strategic interaction.

(1954) Readings in Game Theory and Political Behavior. New York: Doubleday.

An early collection of readings on game theory aimed at political scientists. Collected key papers from the early development of game theory and its applications.

The rest of the summaries focus on Martin Shubik’s extensive contributions to developing game theory and its applications to economics, management science, and other fields from the 1950s onward. He helped establish game theory and simulation/gaming as important tools for analyzing strategic decision-making.

Here is a summary of the cartoon and game theory context described:

The cartoon depicted a scene from the opera Tosca, where the heroine Tosca faces the Chief of Police Scarpia who is holding her lover captive.
The cartoon showed two matrices of “payoffs” representing the benefits of the two choices Tosca and Scarpia each have - whether to keep their promise to each other or break it.
Tosca promises sex favors in exchange for Scarpia ordering blank bullets for her lover’s execution. Scarpia promises blanks but plans to order real bullets.
Neither keeps their promise. Tosca plans to stab Scarpia after he presumably orders blanks, gaining her lover’s freedom and virtue.
Scarpia plans to order real bullets, believing he can then have Tosca and be rid of her lover.
They both end up worse off due to lacking trust and acting selfishly rather than cooperating.
The game structure and choices facing the US and USSR in nuclear strategy during the Cold War were depicted as similar - cooperate by not using nukes, or defect out of distrust and bomb first.
The cartoon and game matrices were used by Rapoport to warn against using game theory to guide dangerous Cold War actions due to the prisoner’s dilemma dynamics.
The Prisoner’s Dilemma poses a dilemma between individual rationality and maximizing individual utility versus achieving the best outcome for both players through cooperation.
The game theorizes two prisoners who can either confess (defect) or not confess (collaborate). Each prisoner aims to maximize their individual outcome but ends up worse off if both defect.
Economists analyze it using a rational economic agent model who seeks to maximize utility. This leads to both players defecting as the equilibrium, even though they would be better off collaborating.
This result that individual rationality leads to a jointly irrational outcome makes economists uncomfortable and challenges key assumptions about rationality and the ‘invisible hand’ thesis that individual self-interest leads to good outcomes.
Allowing learning or cooperation would go against the strict rationality assumption, but sticking to the defection equilibrium also seems problematic. This posed a dilemma for economists in how to analyze and model strategic situations like the Prisoner’s Dilemma.
The prisoner’s dilemma game presents economists with a dilemma - if players act rationally according to self-interest, they will both end up worse off than if they cooperated. But this outcome undermines the assumption that the “invisible hand” leads to efficient outcomes.
Economists have tried to resolve this by broadening what is considered rational behavior to include cooperation. But arguments for things like inherent trust have not been very convincing.
Experiments show that in repeated games, some cooperation does occur, contradicting the theoretical prediction. This challenges the concept of rational economic behavior.
It changed views on whether social and individual rationality are consistent. Game theory showed social outcomes cannot always be assumed to align with individual self-interest.
The prisoner’s dilemma became an exemplar case that demonstrated free markets do not always lead to efficient outcomes, modifying long-held beliefs about the benevolent invisible hand. It demonstrated the “laws of economics” don’t necessarily hold.
It was influential as economists incorporated game theory into their teaching and research from the 1970s onwards to think more critically about rationality and market outcomes.

This section discusses the use of game theory models and typical cases/situations to analyze economic phenomena. It makes the following key points:

Economists reason about specific economic “situations” using game theory models and analyses of typical cases. This provides an intermediate level of explanation between general scientific laws and individual historical explanations.
Popper characterized situational analysis as the method for economic analysis, with models essentially describing reconstructions of typical social situations. This includes the knowledge and environment of individuals within structural relations.
Explanations rely on describing a typical situation, analyzing what is rational within it, and applying a rationality principle that agents act appropriately to their situations. This “packs” the explanatory theory into the model.
Using types/categories of typical situations allows models to provide explanatory breadth beyond single cases but not as broad as general scientific laws. It provides a middle level of coverage for recurrent but not universally applicable instances.
Neglecting the concept of “types” of situations undermines the explanatory power of these modeled analyses, as it depends on typical rather than wholly individual or universal situations.

So in summary, it discusses how game theory and models are used to analyze economic phenomena by constructing and analyzing models of typical situations and cases, providing an intermediate level of explanation.

The author compares microeconomist Paul Milic’s view of “situational determinism” to the concept of “economic behaviouralism”. Both emphasize the status and role of the rationality principle rather than focusing on details of the situation or model.
Narratives play an important role in game theory. They are involved in:

Describing, analyzing and reconstructing the situation/model being represented. Narratives are essentially built into the identity of the game theory model.
Providing flexibility in matching a game theoretical situation to an actual economic situation when applying game theory. Narratives allow for reasoning about the model situation.
Giving explanatory depth when applying models, as the narratives are grounded in thick descriptions of the economic situation, not just informal inferences from the model.

So in summary, the author argues that narratives in game theory go beyond just filling in the middle ground - they are critically involved in defining the model situation and providing robustness to explanations derived from applying the model to real economic cases.

Narratives accompany game theory models and help characterize the model situations. They fill in details like the players’ reasoning and assumptions about rationality.
For the Prisoner’s Dilemma game, the narratives emphasize the rationality assumption that players will maximize individual returns. They also explore ways to allow cooperation but reject these as going against the rationality assumption.
The narratives generate musings about whether binding agreements could be broken, leading Luce and Raiffa to consider revising the game rules or payoffs. But they felt this did not truly solve the dilemma.
Economists recognize a slide between the Prisoner’s Dilemma model and real-world situations it represents, like economic exchanges. Narratives are necessary alongside the payoff matrix to reason about these situations. The model suggests real-world situations need laws/institutions to legislate against defection outcomes.

So in summary, narratives accompany game theory models to characterize situations, assumptions, and help economists reason about analogous real-world cases the models represent.

Narratives play a key role in connecting game theory models to real-world economic situations. They allow economists to smoothly transition between discussing the abstract game model and applying it to a specific case.
Narratives translate the elements of an economic situation (like firms competing in a market) into the structure and payoffs of a game model (like the Prisoner’s Dilemma). They then link the particulars of the economic case back to the typical situation depicted in the game.
Economists use narratives to subtly change or adapt the game model specification as needed to better fit the real-world case. For example, moving from a 2-player to an n-player Prisoner’s Dilemma game.
Narratives are how economists re-describe real economic cases into game theory models while keeping the assumption of rational actors. They also provide a way to reason about and explain what will happen in a particular economic situation.
This process arguably blurs the distinction between analyzing the world using a model versus applying a model to analyze the world. It can impact how economists view and understand the world.
Game theory sees human behavior as guided by “thin rationality” - focusing only on self-interest and outcomes. Psychologists take a broader view of rationality, incorporating context and other human factors.
Economists vary the game situation while keeping rationality constant, exploring different dimensions. Psychologists vary their characterization of rationality while keeping the game situation constant.
Game theory developed taxonomies of different game types to organize modeling efforts. Over time, new categories emerged and the boundaries shifted as interest focused on different aspects of games.
Typical cases or model situations were developed to characterize empirical situations as game scenarios. If a situation did not neatly fit, the game would be modified with new rules or payoffs.
This led to a proliferation of special cases and scenarios that undermined the goal of generalizable theory. While insightful as individual examples, there was no coherent overarching theory that could reliably explain real-world oligopolies and other complex economic situations.
John Sutton (1990) was critical of game theory’s ability to “capture various situations” and argue it was embarrassing because game theory could supposedly “explain” any observed market behavior by deriving it from individually rational decisions.
Sutton bluntly asked if by “explaining everything, have we explained nothing?” He questioned what models excluded.
With game theory able to match many economic situations with multiple candidate models and individual rationality compatible with different equilibriums, game theory lost its ability to provide explanatory power in terms of types of situations (middle-level explanations based on situational analysis).
Explanatory breadth obtained by developing typical cases/taxonomies seemed to have been lost in a “sea of one-off individual cases and anecdotes.”
These critiques from Peltzman, Fisher, and Sutton support the analysis that game theory provides explanations by reasoning about model situations using cases, stories, and aiming for explanatory breadth through typical cases - but this was limited by proliferation of individual cases undermining typicality.
Models and modeling have fundamentally changed economics as a discipline, how economic knowledge is used in the world, and how economists understand the world.
In contrast to Adam Smith’s verbal treatment of political economy in the 18th century, modern economics relies heavily on small mathematical or diagrammatic models to represent different parts of the economy independently.
Models have become the new “working objects” that economists use to describe, theorize, and practice economics. They provide condensed representations of parts of the economic world.
Modeling involves developing networks of simplified models using particular assumptions, examining how the models relate to each other, and drawing inductive inferences from them. It is a new way of practicing economics.
Models are now used as instruments to act in and understand the real world. Economists seek to relate insights from small model worlds to the larger real world. Models provide a way to see small parts of the economic world in isolation yet relate them to the whole.
The modeling community and shared practices are important parts of how models are developed and used in economics. Modeling has become a social as well as technical activity.

So in summary, models and modeling have fundamentally transformed economics as a discipline, how it is practiced, how economic knowledge is applied, and how economists perceive the world. Models are now core objects and tools of economic analysis and policymaking.

In the past century, economic models have become the primary working objects that economists use to do their science. This chapter discusses the nature and role of models more generally.

The first half of the chapter considers what qualities make something a good working object for science. Economic models fulfill these requirements as they are standardized, manageable representations that can be shared and manipulated within the community of economists. Though small and artificial, models can provide sophisticated accounts of economic phenomena. Their representative nature also allows broader insights.

The second half examines the implications of modelling. While individual models seem separate, they are connected through shared assumptions and practices within the economics community. The modelling approach has also shaped how economic knowledge is applied in the real world, changing the way economists view their field from looking at the economy through models to seeing models in the world. Overall, models have created a new engineering mode of interacting with the economy beyond any individual model or use.

The summarized text does not provide any evidence related to what was requested. The passage discusses economic models and their relationship to maps and other representations used in science. It does not mention anything about railway stations, youth hostels, windmills, or evidence from Whitfield (1994/2010).

Small-scale representations are commonly found in various art forms across civilizations, both past and present. They often follow specified conventions and rules regarding what can be depicted and how.
Economic models are a type of small-scale representation that depict economic phenomena in a simplified, abstracted way using mathematical language. However, their small size and simplicity does not necessarily mean the content is also simple or invalid.
Economic models, like poems, compress and articulate accounts of the world in a precise, structured form using specialized languages that can abstract or analogize. Both involve imaginative choices within certain formal constraints.
While economic models are not direct depictions of reality, they serve as “working objects” - abstract conceptualizations that economists use to understand phenomena, make conjectures, develop theories, and analyze behaviors just as other sciences rely on representations.
Comparing economic models to art forms like poems and maps suggests they can be understood not just as toys but as “articulate artifacts” - compressed accounts of the world expressed through an appropriate technical language and form. Their value lies not in capturing all details but enabling theorization within disciplinary standards.

Here is a summary of the key points from ls, Rudwick (1988):

Economic models have served as “working objects” for economists to develop concepts, form hypotheses, solve theoretical puzzles, etc. Their usage has led to development in economic theory over time.
There was a divergence after WWII where statistical modelers focused on empirical descriptions/hypothesis testing while mathematical modelers focused on theoretical accounts and conceptual development.
Models help mediate between general theoretical accounts and specific real-world descriptions. However, there is tension between generalizability and accounting for particular cases.
For models to be useful epistemic tools, they must be manageable in scale but also justifiable as typical/representative of phenomena in the economic world.
Economic models represent “abstract typical cases” or situations to capture features found in specific real-world exchanges, behaviors, etc. and make inductive claims about broader classes.
Models act as “working objects” that economists can investigate and use to gain knowledge with broader inferential scope beyond the specific model.
Economic models function at a generic level, representing typical cases within a class rather than every individual case. This makes them more manageable to investigate but still able to provide insights about the class.
Results obtained within a model can often be generalized, or “induced”, to other cases of the same type via “model induction”. However, this only applies within the scope and assumptions of the particular model.
For modelling to become central to economics, individual models had to fit together somehow even if they were developed separately. This occurred through two key assumptions that provided mathematical rules for modelling, and through models becoming interconnected in different parts of economics in various ways.
The specific way modelling was adopted and practiced in economics, with its own forms, community, etc. helped make it such an influential approach within the field. While modelling has generic qualities common to other sciences, economics developed its own unique version.
Economics used to conceptualize general “laws” but these have mostly disappeared and been replaced by models. Theories have also collapsed into models.
Two key assumptions have replaced the classical economic laws - individual utility maximization and equilibrium tendencies in aggregate systems. These function as shared rules and requirements for model development.
Models are connected at one end to these key assumptions, and at the other end to the goal of developing accounts of economic phenomena. Between is the process of idealization and simplification.
Models fulfill the key assumptions but are otherwise loosely connected and do not fully cover or interlock with each other. They provide customized accounts of phenomena rather than a unified theoretical framework.
Economists have commented that models are like disconnected computer programs written in different languages, rather than bricks forming a coherent structure. In summary, models provide middle-level idealized accounts tethered by core assumptions but not forming a unified system.

The summarized key points are:

Economic models seem like a “patchwork of models” rather than a unified theory, as contrasted with Cartwright’s view of science operating under a “patchwork of laws”.
Modelling became a new style in late 20th century economics, with models tailored to specific markets rather than deviations from a standard model.
Certain models became “keystone models” that connected different parts of economics by being flexibly applied across subfields. Examples include supply/demand and IS/LM models.
Modelling created a network of connections between models beyond just conceptual relationships. It also created a shared community practice and technological skill.
Once modelling became accepted as the proper way to reason in economics, it created a professional commitment to developing all topics as modelling projects. This reinforced modelling as the disciplinary norm.
Certain models like the Edgeworth box and Prisoner’s Dilemma became especially influential by providing conceptual resources, demonstrating theoretical results, or exemplifying implications in a salient way.
There are contingent historical and epistemic reasons why certain models become more central, relating to their ability to reveal typical phenomena or problems in a justifiable way.
Economic models have become central tools in economics that are used not just for analysis and understanding, but also for intervening in and shaping the real world economy. This represents a shift from earlier eras where economics focused more on general laws and principles.
Models provide more detailed and operationalized accounts than general laws, allowing for more direct application to specific economic policies, decisions, and behaviors. They offer “recipes” for acting directly in the world.
Through models and other technical tools, economics has become more of an engineering discipline focused on manipulating and managing the economy, in addition to analyzing it. Government policies influenced by models aim to shape economic outcomes.
However, the ability of economists and models to fully control or remake the economy in their image is limited. A two-way interaction exists, as real-world economic changes also shape developments in economics over time through feedback loops.

So in summary, economic models have become important new instruments that allow economists to more directly engage in intervening in and influencing the real world economy, rather than just analyzing general economic principles, though the influence runs in both directions between economics and the real world.

Fox-Keller (2000) distinguishes between models of things (representing reality) and models for things (with purpose/intervention). The latter differs from the representativeness distinction made earlier in the chapter.
Hacking (1983) discusses the general point about representing and intervening in the world using models.
Models act as “embodiments of purpose” and “instruments for carrying out such purposes”. They can be used directly to shape economic behavior and markets.
Recent econ studies have shown how economists’ models have been performatively used to intervene and remake aspects of markets. Examples include Black-Scholes model in finance and auction/resource models influencing policy.
Popular economics works like Freakonomics translate economic analyses back to everyday language, showing how models can explain phenomena and announcing explanations as a form of surprising common sense.
Models integrate positive and normative elements closer to the practical level compared to earlier economic theories, facilitating their use to directly remake the world according to how economists think it should function based on models.
Economists develop models to represent and understand the economic world. By working with these models through analysis and experimentation, they come to see the world differently than before.
This shift in cognition and perception, enabled by the models, is a prerequisite for economists to act upon the world with their models and change reality. As their actions shape reality, their new ways of representing the world through models end up creating new worlds for all of us.
More generally, adopting new ways of representing a domain, like mathematics for science or abstraction for art, changes how people perceive that domain. For economists, moving to model-based representation similarly changed their view of the economic world.
Initially, economists used models to portray the world. Then, through engagement with models, they began interpreting the real world according to models. Eventually, the models became so ingrained that economists saw models directly in the world, not just as representations.
Photographs were initially thought to distort reality because cameras only have one eye, unlike humans with two eyes. But we have become accustomed to the one-eyed camera perspective.
Similarly, cameras use a sloping back technique so that vertical objects stay upright in photos, rather than converge as they would in a true perspective.
Much philosophical debate has centered on the difficulties of correspondence between models and the real world they aim to represent. This can be treated as issues of inference and how inferences travel between models and reality.
Both the challenges of inference from models to reality, and the use of informal narratives rather than clear inference links, were issues discussed in relation to model usage.
Economists have developed models of the economy that have become so familiar they now see the real world through the lenses of their small models, rather than the other way around.
The chapter acknowledges help from reviewers and readers in developing and refining the ideas presented.

In summary, the passage discusses how photographs and models introduce certain distortions or differences from reality that we become accustomed to, and debates around linking models accurately to the world they aim to represent.

Here is a summary of the provided sources:

Several sources discuss Edgeworth’s 1881 contract curve model for analyzing exchange, including its use by later economists like Cournot, Edgeworth, Humphrey, Leontief, Marshall, Pareto, and Scitovsky.
Sources discuss the development of economic man and homo economicus models from classical political economy through Jevons, Menger, Knight, Mill and post-neoclassical economics.
Papers examine the use of demand and supply curve diagrams by economists like Jenkin, Mangoldt, Marshall and Rau.
Frisch’s work developing prototype and sample survey methods is discussed.
Sources analyze game theory and the prisoner’s dilemma, including the development of game matrices and early applications in economics.
Papers explore the role of models, experimentation, and simulation in economics, especially related to Keynesian ideas and work by scholars like Arrow, Debreu, Fisher, Hicks, Samuelson, Shubik, and Simon.
Broader methodological discussions address topics like idealization, abstraction, conceptualization, concretization, and the cognitive and community aspects of model-building in economics.
The numbers represent various aspects of models and modeling: 129 refers to small-scale models or ‘miniatures’, 150 refers to ideal types, 219 refers to models conceived as analogies, etc.
Terms like narrative, semi-cooperative, and exaggeration refer to different types or uses of models.
Locations or people mentioned likely relate to specific historical examples or cases of modeling, like Gatcomb Park farm, Weber, Nash, etc.
Modeling functions or aspects discussed include exemplifying theory, experimental design, explanatory breadth and depth, generalization from examples, and more.
Particular model examples mentioned include IS/LM, Lotka-Volterra, mechanical balances, cobweb models, and more.
Overall it touches on many different facets of models and modeling like their historical development, uses in different contexts, representations versus reality, idealizations, inference from examples, and roles in theories. But it does not provide a clear narrative or develop any points in detail.

So in summary, it lists and references various modeling terminology, examples, functions, and aspects but without much context or explanation of the connections between the individual references.

Here is a summary of the key points from the sections provided:

The sections discuss modelling as a method of reasoning and inquiry in economic science. They examine how modelling practice has changed over time, with economics moving towards more formalized and quantitative modelling approaches using objects like models.
Modelling is analyzed as a style of practical reasoning that uses formal rules and resources. Different forms that models can take are explored, as well as how models are made and used to reason about economic situations.
Ricardo’s numerical model farm from his Principles of Political Economy and Taxation is examined. It is argued that Ricardo constructed a model farm using numbers from experimental farming accounts to represent and reason about distribution questions. This constituted constructing the first numerical economic model.
Issues of world-making and representation in models are discussed. The development of the Edgeworth box diagram is analyzed as an example of how models can generate new ways of imagining and visually representing economic concepts and spaces.
Questions are raised about whether models translate the world or generate entirely new conceptual worlds. The differences between how artists and economists visualize spaces are also examined.
In general, the sections analyze modelling as a method that has changed economic science by allowing different forms of representation, reasoning and inquiry about economic concepts and situations. Modelling is positioned as a key part of the disciplinary practice of economics.

Here is a summary of the key sections from “Types, Idealization, and the Art of Caricature”:

Section 2 characterizes the classical economists’ conception of “Homo Economicus” as a rational, self-interested actor.
Section 3 discusses Max Weber’s idea of “ideal types” which abstractly characterize entities, and Carl Menger’s view of the economy as involving purposeful human action.
Section 4 examines William Jevons’ conception of the “calculating man” who rationally calculates costs and benefits.
Section 5 looks at Frank Knight’s characterization of decision-making as akin to a “slot machine” involving uncertainty.
Section 6 analyzes how “rational economic man” became a dominant simplified representation/role model in economics.
Section 7 discusses caricatures and the processes of abstraction and idealization involved in model-building.
Section 8 examines how the characterization of “economic man” has changed over time as economics has undergone “de-idealization”.

In summary, the chapter traces the evolution of simplified representations of human behavior in economics from early conceptions to the dominance of the rational actor model, and discussions of caricature, idealization and abstraction involved in model-building.

#book-summary

The World in the Model - Morgan, Mary S_