What this covers

This is an EconTalk conversation in which Russ Roberts questions Ricardo Reis of Columbia about how Keynesian ideas are taught and used in research, and what the 2009 crisis shows about the field. It begins in the undergraduate classroom with Y = C + I + G. Reis and Roberts agree that this is an ex post accounting identity that says little about causation until each component has a behavioral equation behind it. The talk then moves to the graduate-level New Keynesian model and argues that its three equations share a core mechanism with IS-LM: a real interest rate shaping the choice between consuming today and tomorrow.

The middle of the conversation is about prices. Reis defends sticky prices as necessary for output to fall below potential. He locates the stickiness in the cost of processing information, citing a study of a Midwestern parts manufacturer, and not in the cost of reprinting menus. He also says that Miles Kimball was right that the New Keynesian and new monetarist models are close to the same model. He notes that dropping IS-LM from teaching loses the paradox of thrift. Roberts also brings in Milton Friedman's information-based account of money's short-run effects.

The history and policy sections are the most contentious. Reis lists lessons from the Great Depression: the dangers of fixed exchange rates and the gold standard, the damage from protectionism, and the cost of letting banks collapse on a large scale. He credits Roosevelt with helping end the Depression but suggests the NRA and cartel policies slowed recovery, citing Lee Ohanian. On the present crisis, he rates monetary policy as adequate, with Ben Bernanke acting in line with theory. He puts the real gaps elsewhere: thin empirical evidence on fiscal policy, no legal way to resolve large banks, and bank decisions driven by political feasibility. Near the end the talk turns to why macroeconomics still predicts poorly and whether it counts as a science.

Sharpest takeaway

Reis argues that IS-LM remains worth teaching as a general-equilibrium scaffold, that accounting identities like C+I+G mislead without behavioral models, and that macroeconomics' weakness lies in thin fiscal and banking evidence and poor prediction rather than in monetary theory.

  • IS-LM is the only general-equilibrium exercise in undergraduate textbooks, and it maps onto the three-equation New Keynesian model
  • Y = C + I + G is an accounting identity that cannot establish causation without models of each component
  • Monetary policy has solid theory and data, while fiscal policy has almost no empirical grounding and banking knowledge is lacking

The claims · ranked60 claims · weighted by value

This asset isn't compiled yet

You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.

0.81

Macroeconomists have models of why government spending matters and models of how spending should be set optimally, but almost none model how government spending is actually set through the political process, so the variable G is treated as exogenous.

factualhigh valueestablishednovelty 3/4durability 3/4· Russ Roberts

“in most macro class is the role of G which is a rather simplistic letter to describe the complexity of the political process it isn't really modeled”

0.81

The FDIC cannot resolve most of a modern bank's problems because it lacks legal power over bank holding companies and non-bank parts of the balance sheet; even if it took over Citibank, most of its balance sheet sits outside the bank in the shadow banking system.

factualhigh valueestablishednovelty 3/4durability 3/4· Ricardo Reis

“if you let FDIC go into Citibank you'll make very little of a difference because most of the Citibank balance sheet is outside the bank”

0.81

Empirical knowledge of fiscal policy is very thin: monetary policy at the zero lower bound has a substantial empirical literature from Japan and quantitative easing, while fiscal work such as Romer and Romer and Valerie Ramey's studies only scratches the surface and rarely distinguishes between different kinds of spending.

factualhigh valueestablishednovelty 3/4durability 3/4· Ricardo Reis

“when it came to fiscal policy though empirical evidence in the end it's just very very small”

0.76

Fixed exchange rate regimes, including the gold standard, were a major reason for the worldwide Great Depression, and economists have understood the dangers of fixed exchange rates better since the 1970s shift to floating rates and the 1980s and 1990s currency crises.

causalhigh valueestablishednovelty 2/4durability 4/4· Ricardo Reis

“something like the gold standard may have been one of the big reasons why we have such a great depression particularly our worldwide Great Depression”

0.75

The first lesson from the Great Depression is that letting banks collapse on a large scale is very serious, because it destroys information and disrupts the channeling of funds from investors to borrowers, so policy should keep the flow of credit and the knowledge of which projects are good alive rather than saving particular institutions.

normativehigh valueestablishednovelty 2/4durability 3/4· Ricardo Reis

“it's very important or it's very sorry it's very serious and bad to let banks collapse because we win banks collapse at a very large scale there's a great loss of information”

0.75

In the old Keynesian framing, a fall in consumption reduces output through a shortfall in aggregate demand, while in the new framing, output falls today only relative to the future because saving shifts resources into investment that raises output tomorrow, so the question becomes present versus future rather than C versus I.

causalhigh valueestablishednovelty 2/4durability 3/4· Ricardo Reis

“in the new we really think in terms of present versus future is output in consumption lower today relative to what it will be in the next period”

0.74

Milton Friedman's monetary history account, with Anna Schwartz, argued that money supply changes have short-run real effects because firms cannot distinguish economy-wide inflation from a specific rise in demand for their product, but in the long run money only changes inflation; this attacked both sticky prices and the Phillips curve.

definitionhigh valueestablishednovelty 1/4durability 4/4· Russ Roberts

“Milton Friedman in 1962 when he with Anna Schwartz wrote the monetary of the United States it really began as the beginnings of a counter-revolution”

0.74

Protectionist responses to a large economic shock turn a deep recession into a Great Depression, as happened in the 1930s with trade barriers, and the value of free trade was more widely recognized in the last 20 to 30 years.

causalhigh valueestablishednovelty 1/4durability 4/4· Ricardo Reis

“reacting threw a big shock or a big depression for an output through protectionist measures is really a sure path to turn a deep recession into a Great Depression”

0.73

The New Keynesian model and the new monetarist model are essentially the same model: Miles Kimball first made the point that both justify monetary policy having real effects through price adjustments, differing mainly in how large they think the real effects are.

causalhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“miles Kimball was the first one to make this point that the new Keynesian model could really be described as the new monetarist model”

0.73

The most plausible source of price stickiness is not unchanged printed prices but the costly processing of information: firms must acquire, absorb, and process data and then bargain internally to decide what price to set, so they respond slowly to economic news even when changing a price is cheap.

causalhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“the most plausible source for why prices don't adjust is exactly a friedman lucas information type story”

0.73

Franklin Roosevelt probably helped end the Depression, but some of his policies, such as the NRA and the expansion of cartel and union power, probably stalled economic growth; Ohanian's work argues this contributed to the length of the Depression.

causalhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“Roosevelt probably did get us out of the depression and deserves his rightful place in history for that also made some mistakes one of which was probably something like the NRA”

0.73

An accounting identity such as Y = C + I + G has too many unknowns to be useful by itself; it becomes powerful only when combined with behavioral equations for consumption, investment, and government spending, which together with market-clearing conditions describe the economy's mechanisms and trade-offs.

causalhigh valueestablishednovelty 2/4durability 4/4· Ricardo Reis

“if you just stare at y equals C plus I plus G the accounting identity it's hard for me to see that you really make much progress”

0.73

Macroeconomics is good at isolating mechanisms and telling stories about data but still fails at prediction, and this failure is repeatedly exposed by each recession or economic event.

factualhigh valueestablishednovelty 2/4durability 4/4· Ricardo Reis

“the next step is being able to predict what will happen is one where I think we're still failing”

0.72

A study by Zbaracki and coauthors at a large Midwestern parts manufacturer found that the costs of price setting, including meeting time, catalog printing, and customer communication, amounted to roughly 1 to 2 percent of net operating margin, and that most of it went to gathering information and agreeing internally.

factualhigh valueestablishednovelty 3/4durability 2/4· Ricardo Reis

“there were things like one or two percent of net operating margin and most of them have to do essentially with gathering information”

0.71

The New Keynesian three-equation model used in graduate teaching is conceptually the same as the IS-LM framework in its core mechanism: both center on a real interest rate affecting consumption and saving decisions, with the new version framed as an intertemporal trade-off between consuming today and consuming tomorrow.

definitionhigh valueestablishednovelty 2/4durability 3/4· Ricardo Reis

“in the new model it's really very similar in that it's really about the real interest rate and how it effects how do I want to consume today versus tomorrow”

0.71

Sticky prices are needed in the Keynesian story to explain why output can be lower than potential: prices set too high from a prior period can stay high, and sticky nominal prices allow monetary expansion to change real variables rather than just inflating prices, with money otherwise acting as a veil.

causalhigh valueestablishednovelty 2/4durability 3/4· Ricardo Reis

“if the price were just to adjust one to one with that amount of money then nothing would really change we just have money as a veil”

0.71

Price stickiness should be defined as prices failing to respond to current economic news, not as prices being literally unchanged; the cost of acquiring information has fallen sharply online, but the cost of absorbing and interpreting it remains as high as ever.

definitionhigh valuecontestednovelty 3/4durability 4/4· Ricardo Reis

“when I say sticky prices I do not mean prices that are unchanged because we're written a piece of paper but I mean the fact that you are not changing your price in response”

0.70

The paradox of thrift appears in the old IS-LM model, where an attempt to save more reduces output, but it does not show up in the New Keynesian model, so dropping IS-LM from teaching loses the paradox of thrift as a teaching idea.

causalhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“the paris of thrift shows up in the old one but not in the near one”

0.70

The US lacks a legal mechanism for an orderly resolution of large banks comparable to Chapter 11 for firms, so regulators cannot let major banks fail in a controlled way; these are sticky laws, not sticky prices.

factualhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“we just can't do because we do not have legal institutions in the u.s. that will allow us to do an orderly bankruptcy”

0.70

Policy decisions on banks during the crisis were shaped more by political feasibility and legal constraints than by sound economics, because the US lacked a legal chapter 11 equivalent for banks, so an orderly resolution of failing banks was not possible.

causalhigh valuecontestednovelty 3/4durability 3/4· Ricardo Reis

“there's just really really some right things like letting some banks bail which we just can't do because we do not have legal institutions in the u.s.”

0.69

Recent US banking policy decisions, including TARP and related Bush-era and subsequent moves, were driven mainly by political and marketing constraints rather than sound economics, and policymakers still avoid letting large banks fail because of political fear.

causalhigh valuecontestednovelty 3/4durability 2/4· Ricardo Reis

“a lot of the current recent policy decisions have been made very much subject to a series of constraints”

0.69

Despite massive expansion of the monetary base by Ben Bernanke, which Roberts notes has roughly doubled in the last year, monetary policy has been adequate given economic theory and models, and monetary policy failures are less significant than failures in knowledge of banking and fiscal policy.

normativehigh valuecontestednovelty 3/4durability 2/4· Ricardo Reis

“monetary policy is not one where I've seen big failings of last couple of years I've seen failings our knowledge of the empirical effects of fiscal policy”

0.69

Economists lack a good empirical model of government spending in the current crisis, so proposals like raising stimulus spending by a set amount treat the composition of spending as irrelevant, which Reis calls a very embarrassing fact.

normativehigh valuecontestednovelty 3/4durability 2/4· Ricardo Reis

“it's a very embarrassing fact G has been it's been proposed that G go up 787 billion dollars”

0.68

Fiscal stimulus can boost output, a lesson Keynes argued very persuasively and that monetarists later questioned, but the quantitative size of fiscal effects is highly uncertain, and recent work by Gauti Eggertsson shows fiscal spending creates expectations of future inflation and lowers real interest rates even at the lower bound.

causalhigh valuecontestednovelty 2/4durability 3/4· Ricardo Reis

“fiscal standing essentially is just creating expectation of future inflation as well as being able to affect real interest rates even though we're at the lower bound”

0.68

Macroeconomics went through a revolution around 25 to 30 years ago, comparable to relativity in physics, that changed the basic structure of models; in the mid-1980s macroeconomists could answer almost no policy questions, while 20 years later they could give precise answers about monetary policy effects.

factualhigh valuecontestednovelty 2/4durability 3/4· Ricardo Reis

“macroeconomics went through a real revolution what about 25 to 30 years ago it was a real change in how even the basics of the model get changed”

0.68

Economists failed to predict the housing market collapse and the severity of the current recession; almost no one in academia had a systematic consensus that the crisis was coming, and Ben Bernanke, a first-rate monetary economist, did not anticipate it.

factualhigh valueestablishednovelty 2/4durability 3/4· Russ Roberts

“no one anticipated this problem a lot of people predicted that not a lot a few people predicted that the housing market might collapse”

0.68

Christina Romer's work using exogenous changes in US fiscal policy, unrelated to the business cycle, is a smart use of modern econometrics that identifies causal effects, but it is only the first study and needs ten to a hundred more before the question is settled.

normativehigh valueestablishednovelty 2/4durability 3/4· Ricardo Reis

“what it does precisely is try and look in the history of the u.s. when is it that we had changes in the fiscal policy that were driven by exogenous things”

0.68

Reis agrees with Roberts that the simple C+I+G undergraduate story is an ex post accounting identity that is true but misleading about causation, because output comes from the bottom up through wealth creation and innovation rather than from a top-down apportioning of spending.

causalhigh valuecontestednovelty 2/4durability 3/4· Ricardo Reis

“it's just imposing an ex post accounting identity on it is that a fair criticism”

0.68

Sticky prices were plausible in the 1960s and 1970s because wages were set by long-term union contracts and prices were costly to repaint, but in the 21st century unions are under 10% of the private workforce and most prices are set by barcode systems that can be changed instantly, so those mechanisms no longer explain stickiness.

factualhigh valuecontestednovelty 2/4durability 3/4· Ricardo Reis

“now we're an economy in the 21st century where unions are less than 10% of the private sector workforce most prices are set by a bar code”

0.62

Many current macroeconomists argue the Great Depression lasted so long because the US never tried true deficit spending, since Roosevelt was too worried about balancing the budget, and that the Depression lesson is that deficit spending staves off depression.

causalhigh valuecontestednovelty 1/4durability 3/4· Russ Roberts

“a lot of current macro folks will argue that the reason the depression lasted as long as it did in the 30s was because we never really tried the Keynesian remedy of true deficit spending”

0.61

The Friedman-style monetarist and New Keynesian positions differ mainly in how effective they think monetary and fiscal policy are at controlling the business cycle: monetarists emphasize imperfect information and policy weakness, New Keynesians argue policy can control output.

causalhigh valuecontestednovelty 2/4durability 3/4· Ricardo Reis

“the Milton Friedman view would emphasize as you said imperfect information stories”

0.57

The fiscal multiplier debate is at a crude level: proponents cite the Great Depression ended by World War II spending, critics argue wartime prosperity was an illusion because of mismeasurement and that the war only benefited weapons makers and the army, and Japan's infrastructure spending in the 1990s is cited both ways.

factualhigh valuespeaker onlynovelty 3/4durability 3/4· Russ Roberts

“the crumbing level says well we know fiscal stimulus works look at the Great Depression it was ended by the war”

0.57

Much of the confusion in public and policy debates about government spending arises because participants focus on one equation and omit their underlying models; two people can accept the same accounting identity yet reach opposite conclusions because they hold different models of where consumption and government spending come from.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Ricardo Reis

“they're not disagreeing with the accounting identity they really know their model for what's going on behind with the engine”

0.56

The evidence on the fiscal multiplier ranges from zero to 1.5, with proponents of spending citing the high end, others finding it near zero or even negative, and no decisive evidence available to resolve the disagreement.

factualhigh valuecontestednovelty 2/4durability 2/4· Russ Roberts

“the multiplier which ranges you know it's tricky to how you define the multiplier when it ranges from zero to one point five seven”

0.55

Economists often describe partial derivatives, focusing on one channel of an effect such as government spending through aggregate demand, even when a total derivative with many other effects exists; criticizing them for ignoring other effects misreads a matter of focus.

normativehigh valuespeaker onlynovelty 3/4durability 4/4· Ricardo Reis

“they're always essentially taking trying to describe partial derivatives even if there's a total derivative behind it”

0.54

Macro education has a three-stage structure: models of why government spending matters, then models of how it should optimally be set, and then the missing third stage of how it is actually set; the omission is partly a matter of running out of time and partly because the third stage is too hard to model.

factualhigh valuespeaker onlynovelty 3/4durability 3/4· Ricardo Reis

“we then never get to the third stage which would be how is she actually set”

0.54

Information acquisition costs have fallen sharply with the internet, but the costs of absorbing and interpreting information within a firm, including internal bargaining, remain as high as ever, so the bottleneck is processing rather than gathering data.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Ricardo Reis

“the cost is absorbing it in terms of figuring out what does it all mean are as high if not higher than ever”

0.54

Policymakers say they avoid banking collapses to preserve information flows, yet they have let almost no banks fail, so they are swinging too far, keeping in place decision makers who made bad decisions.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Russ Roberts

“in the name of that we've let almost no banks fail and as a result that knowledge that you're talking about”

0.54

Macroeconomists using econometrics on fiscal policy are often doing faith-based work that confirms their biases, since in complex systems the anecdotal narrative fallacy persists even when dressed in econometric form.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Russ Roberts

“I'm very drawn to this faith based econometrics people doing econometrics that just confirms their biases”

0.53

Fiscal policy's evidence gap reflects a lack of research focus over the last 20 to 30 years, not an inherent flaw of economic science, while monetary policy progressed rapidly because it was a hot research area with good theory, data, and estimates.

causalhigh valuespeaker onlynovelty 3/4durability 2/4· Ricardo Reis

“it's more that we've only been doing this for twenty thirty years fiscal policy has just not been a very exciting area”

0.52

Reis argues IS-LM should still be taught to undergraduates even though it has disappeared from active research for about 20 years, because it is the only undergraduate textbook instance of general equilibrium, forcing students to see feedback between the money market and goods market until both clear.

normativehigh valuespeaker onlynovelty 2/4durability 3/4· Ricardo Reis

“I am is actually the only instance of general equilibrium in the undergraduate textbooks”

0.52

In the absence of empirical evidence on fiscal policy, public debate becomes a set of economists pushing their pet theories or ideologies, which is muddled and resembles an academic seminar rather than a debate grounded in data.

normativehigh valuespeaker onlynovelty 2/4durability 3/4· Ricardo Reis

“what you see is people essentially talking about either their pet theory or their pet ideology”

0.47

IS-LM should be understood as a way of organizing thinking that handles macro complexity, not as truth or empirical science, and its teaching value lies in that organizing role.

normativehigh valuespeaker onlynovelty 2/4durability 3/4· Ricardo Reis

“it's a way of organizing our thinking it's not truth it's not right it's not empirical science it's a framework”

0.44

Because the interest-rate and investment mechanisms in undergraduate IS-LM are essentially the same as those used in research macro, teaching IS-LM to undergraduates involves little sacrifice of substance.

normativehigh valuespeaker onlynovelty 2/4durability 2/4· Ricardo Reis

“we do in research macro are also present essentially very similar mechanism in the undergraduate macro so it's actually not so much of a sacrifice”

0.44

Modern econometric methods, such as Romer and Romer's use of exogenous fiscal policy changes in US history to identify causal effects, are a smart way to estimate fiscal multipliers, and with enough studies, similar to the hundreds of studies on monopoly, the field could reach reliable conclusions.

forecasthigh valuespeaker onlynovelty 2/4durability 2/4· Ricardo Reis

“I think it gives us a very good picture of some things what I was complaining about was not that that study was inadequate”

0.43

Keynes's ideas are pervasive in mainstream undergraduate macroeconomics textbooks, present in nearly every macro chapter in the way Adam Smith's ideas are present across economics teaching, often appearing as opposition to or response to Keynesian ideas even in real business cycle discussions.

factualestablishednovelty 1/4durability 3/4· Ricardo Reis

“they're really pervasive so in some ways they're as present as are the ideas of Smith or Ricardo”

0.35

In the old IS-LM model, the intertemporal dimension was ignored and the interest rate affected output mainly through investment, so a higher rate lowered investment and output, and this investment channel drove much of the model's dynamics.

causalestablishednovelty 1/4durability 3/4· Ricardo Reis

“higher interest rate is going to lower investment in that lower output that's in the old model”

0.35

The classical model has flexible prices and wages, no information problems, and no Phillips curve, so monetary policy is completely ineffective; the New Keynesian model instead holds that aggregate demand policy can affect output, which can be too low or too high.

definitionestablishednovelty 1/4durability 3/4· Ricardo Reis

“the classical model is one in which there's flexible prices and wages there are no information problems”

0.32

Austrian business cycle theory, which holds that interest rate distortions misallocate capital into unusable forms such as machines, shares a structure with sticky price theories in that distortions from prices that do not adjust cause misallocation of resources, though here the relevant prices are goods prices.

definitionfringenovelty 2/4durability 2/4· Russ Roberts

“in the Austrian theory of a price the interest rate is going to distort a bunch of decisions and lead to both a loss of output and cycles”

0.29

Economists broadly agree that cartels which enrich one group at the expense of the rest do not create wealth, and that shifts in the money supply and monetary aggregate collapses are harmful, yet these consensus lessons are easy to misapply in the current crisis.

normativeestablishednovelty 0/4durability 3/4· Russ Roberts

“every economist virtually every enormous consensus that cartels that enrich one group at the expense of the rest economy don't create wealth”

0.29

The standard undergraduate story is that a fall in aggregate demand, from consumption C plus investment I plus government spending G falling, creates a gap between actual and potential output, which creates room for fiscal or monetary stimulus.

causalestablishednovelty 0/4durability 3/4· Russ Roberts

“if one of those goes down such as C if consumers start to spend less we're going to have the beginnings of a Keynesian business cycle theory”

0.28

Reis's own research on imperfect information as the source of price stickiness places him in a new monetarist camp, yet his work is categorized as New Keynesian because it is framed against the classical approach, which the labels do not reflect.

factualspeaker onlynovelty 2/4durability 2/4· Ricardo Reis

“my own work is all about imperfect information stories of why prices don't move enough”

0.27

Reis has never seen convincing evidence that the paradox of thrift is an important empirical fact, and also has not seen convincing evidence that it is false; he considers it an important open issue.

factualcontestednovelty 1/4durability 2/4· Ricardo Reis

“I have never seen very convincing evidence that that's the case but neither that it's the opposite”

0.26

Paul Krugman's argument that the stimulus should be about two trillion dollars rather than one trillion is based on a partial derivative of government spending through the aggregate demand channel, ignoring the other effects of spending that depend on what kind of spending it is.

causalspeaker onlynovelty 2/4durability 2/4· Russ Roberts

“he isn't worrying about what the truly extra trillion is spent on he just wants more spending”

0.26

The sandwich shop outside Reis's office has not cut prices during the recession even though local demand is falling, because the owner is still trying to figure out what is happening, which illustrates sticky information rather than sticky prices printed on signs.

causalspeaker onlynovelty 2/4durability 2/4· Ricardo Reis

“why is it that the guys selling sandwiches outside your office even though we can easily write a different price”

0.25

IS-LM is valuable for teaching because it lets students see how monetary and fiscal policy affect the economy, providing an intuitive apparatus that is needed before students can understand more advanced models like the Barro-Gordon model of inflation bias and Ricardian equivalence.

normativespeaker onlynovelty 1/4durability 3/4· Ricardo Reis

“it's an important foundation for the students to be able to see that so that then we can vary to get to the policy questions”

0.25

Economists know the general principles of banking policy, but applying them is hard in practice because real banking systems are complicated, so the scientist's job is to clarify idealized mechanisms and let policymakers decide.

normativespeaker onlynovelty 1/4durability 3/4· Ricardo Reis

“in numerators and general principles but when it comes to the real action they become hard to apply”

0.21

Policy debate on banking and deficit spending is not science but ideology, with Nobel laureates on each side calling each other names, so the exercise is demeaning.

normativespeaker onlynovelty 1/4durability 2/4· Russ Roberts

“to my mind it's just ideology it's not science and we are it's a very demeaning exercise”

0.21

Prices are in fact falling in the current downturn: mail-order retailers are sending many discount emails, so prices do adjust, just slowly and over time.

factualspeaker onlynovelty 1/4durability 2/4· Russ Roberts

“a lot of price scores are falling they're getting a lot more emails from the mail-order places I use offering me discounts”

0.21

It is strange that 70 years after the Great Depression and 73 years after Keynes's General Theory, the policy prescriptions for the current crisis, increasing money supply and spending more, are the same ones that would have been chosen in any of the last 40 years.

normativespeaker onlynovelty 1/4durability 2/4· Russ Roberts

“the solutions for the mess we're in seemed to be the same ones that we would have chosen any time in the last I would say 40 years”