Saturday, October 31, 2020

Rhetoric of economic policy -- Biden plan analysis

Last week saw four interesting statements by economists regarding the economic effects of Biden economic plans. 

My focus will be "An Analysis of Vice President Biden’s Economic Agenda: The Long Run Impacts of Its Regulation, Taxes, and Spending" by  Timothy Fitzgerald, Kevin Hassett, Cody Kallen and Casey Mulligan, a 50 page report. (Yes, hosted by the Hoover Institution, my employer). The Wall Street Journal gave it major coverage in its editorial page, offering a thoughtful summary.   

I  contrast that piece with  a letter  signed by 13 Nobel-Prize winning economists  endorsing Biden's economic policies. A separate open letter  signed by 1072 regular economists wrote, and a similar Economists for Trump letter.   

I am a bit late to the game, as it took me a while to read the weighty Hoover report. However, unlike letter writers, I have no illusions that my opinions will sway the election. 

And, if the polls are right and Biden wins, the question of just what Biden's policies will be, and their economic effects, will have perhaps greater resonance after the election than in the Biden vs. Trump choice of the election.  As they formalize, debate, and institutionalize the plans, surely quantitative analysis of the likely outcomes will matter.  

I highlight this report not because of its contribution to the issue of the day. This report marks a dramatic innovation in rhetoric, how economists analyze political plans. The authors really have started a revolution in policy analysis. This was evident in their previous work at the CEA, but the report highlights it. 

*****

The report. 

The deep innovation: This is entirely, and appropriately, a neoclassical analysis. This report shows you how to do serious, quantitive, applied, large-scale detailed and transparent incentive-based analysis. 

(I use "incentive-based" as a clearer and less charged word than "neoclassical" these days. It gets to the central point.) 

This report puts the neoclassical growth model at the center of policy analysis, rather than the simple Keynesian ISLM model. And that's exactly appropriate for permanent long-run policies, not short-run get out of a depression policies. 

This is a Big Deal. Past the current election, the report sets out a machinery that can revolutionize policy analysis. (Granted, the report is just the latest in a development of this effort. The recent Economic Reports of the President which the authors cite and build on are precursors.) But that machinery is the big point of this post, and an invitation of a project for others to emulate -- no matter whose policies look good or bad by it. (Trade and immigration restrictions will not look so good by these methods.) 

By contrast, most analysis of "president x" plan, when you get under the hood, amount to this: Candidate X proposes to spend $1 trillion additional dollars. Multiply by 1.5 to deduce that this will "stimulate" the economy and raise GDP by $1.5 trillion dollars, or 7.5% ($20 Trillion GDP). Divide by 150 million employed to conclude that candidate X will "create" 10 million new jobs.  Yes, this sort of hydraulic Keynesianism pervades current policy analysis. 

Just how higher taxes don't reduce such stimulus as much takes some skulduggery. But most analysis of taxes is entirely on their redistribution effect, who pays a bit more, who pays a bit less, in enormous detail, without the slightest concern over who actually bears the burden of taxes -- no "corporations" do not pay a cent of taxes, in the end workers, consumers, or investors do -- and zero treatment of disincentives and behavioral responses. 

Most analysis present results of a black box, unverifiable. The report is by contrast transparent, reproducible, and tells the story of its economic mechanisms. 

An example of incentive-based analysis, and explaining the mechanisms: what are the effects of the corporate tax? Not "corporations pay their fair share!" as we know by simple addition corporations pay nothing. 
Any empirically-grounded estimate of the effects of taxes (or business regulation) must confront the fact that national average after-tax returns on capital have been fairly constant over long periods of time and across a broad cross section of countries despite large differences in rates of capital taxation. [For example, calculations of after-tax capital-rental rates for the twentieth-century U.S. find rates of about 8 percent per year at the beginning of the century – when corporate and personal income taxes were unconstitutional – and also about 8 percent at the end of the century (C. B. Mulligan 2002). For cross-country patterns, see Caselli and Feyer (2007).] This is strong evidence that, in the long run, the owners of capital have close substitutes to investment in the businesses of a specific country. Faced with a high tax rate on their capital income, eventually they reduce investment in the jurisdiction until pre-tax profits are high enough to compensate for the high tax rate. In effect, workers and consumers eventually pay for capital-income taxes rather than the capital owners who are legally liable for the tax.
(They acknowledge criticisms of this view.) In short, tax capital and you get less of it -- a lower capital stock, until the remixing machines are more productive, enough to pay the tax. With less machines, we get less GDP. And with less GDP, workers are less productive too, and so wages decline. 

So why do jobs decline? If we're not Keynesian, the labor market clears, so one might think wages decline, but not number of workers. Indeed, the decline in productivity 

reduces the marginal product of labor, and hence wages. However, such a wage change has little effect on overall employment or hours worked because of opposing income and substitution effects on labor supply. 

Translation: if you get a lower wage, the incentive to work is lower as each additional hour gets you less money. But you're also poorer, and poorer people work more. We usually think these effects roughly offset so lowering after-tax wages does not hurt jobs. 

People work less overall because the Biden agenda also increases marginal tax rates on work as it redistributes to the unemployed, low-income households, and producers in protected industries....

Most of the additional labor (or consumption) taxation comes from changing rules for subsidizing ACA plans. Regulation generates a labor wedge to the extent that it reduces competition.... 

Most of the additional labor (or consumption) taxation comes from changing rules for subsidizing ACA plans. Regulation generates a labor wedge to the extent that it reduces competition.

These changes give the substitution effect without the income effect, driving people to work less. 

But to the point, let us examine less the answer than the audacious question. The report does not analyze demand-side "stimulus," direct government employment, jobs programs or other bits of economic magic, or at best short-run analysis. It gets employment effects from labor supply, circumstances under which people choose not to work as much. This is entirely appropriate for long-run questions. 

The quantification of cost of regulations is another analytical innovation that should become routine. They approach climate regulation in a straightforward way: The Biden-propsed climate policies amount to producing energy in a less economically efficient way, while raising taxes to subsidize the price of energy. There's no getting around it, less efficient ways of producing energy reduce GDP, and taxes distort economic activity. That "green" energy requires so many "jobs" is a cost, not a benefit. They quantify this channel. 

Regulation is much harder to evaluate than standard policies that fit easily in the neoclassical growth model or even Keynesian demand models. One of many innovations of this report is a blueprint of how to do it. 

Admittedly, 

 we do not assess the benefits those policies might deliver aside from rough estimates of the tonnage of carbon abatement. Climate change is real.  

But this is a general failing. Where is the estimate of how many tons of carbon the whole green new deal reduces, how much global temperatures decline, and how much GDP is thereby saved in 2100? Not in the Biden plan! (At least as far as I have seen - I welcome numbers if they are there.)

Though it seems long, the report has a strictly limited scope. It assesses a few tax, regulation, and energy proposals, but it does not try to be a full evaluation of the Biden plan. As such it is extremely conservative, in the other sense of the word -- it simply ignores thousands of policies where it might have easily found economic damage. On the other hand, it is much more detailed than the typical academic article.  This is the novelty: serious, medium-scale reproducible incentive-based analysis. 

I would have gotten lost, and learning how to boil policy down to something manageable is an important lesson. The energy part of the Biden Plan alone goes on and on. A few random quotes: 

Ensure that environmental justice is a key consideration in where, how, and with whom we build – creating good, union, middle-class jobs in communities left behind, righting wrongs in communities that bear the brunt of pollution, and lifting up the best ideas from across our great nation – rural, urban, and tribal...

Mobilizing the next generation of conservation and resilience workers through a Civilian Climate Corps. 

 ...comprehensively address the most pressing, intersectional environmental justice issues and hold polluters accountable. ... Biden will establish a new Environmental and Climate Justice Division

"Intersectional environmental justice" issues? How do we analyze that? Go on and read the Democratic Party platform, and the total number of proposals is overwhelming. How do you deal with that? 

You don't. The report limits its scope to a medium-sized number of quantifiable policies. This is a sweet spot -- much more detailed than typical analyses, but still at a manageable level to be comprehensible. 

The result is remarkably benign.  A cocktail party conversation of conservative economists, who have worked for Trump, asked how will confiscatory taxation, green new deal, and regulatory onslaught will affect the economy, might respond in end-of-western-civilization tones -- similar but opposite to the uninformed enthusiasm expressed by the letter writers, documented below. 

The report's answer is only a loss of 8.5 percentage points of GDP, and 3 percentage points of employment (jobs). That's only a 1 percentage point drag on growth per year over two administrations, a four year delay in the march of prosperity. The UK's GDP per capita is 32% lower than the US ($42,944, vs. $62,795) so they're only accusing Biden of taking us 1/4 of the way to the UK, and many people might not consider that a disaster.  Herkenhoff, Ohanian, and Prescott find that land use regulation alone costs the US 12 percentage points of income. 

In sum, these are very modest claims of damage -- which I think raises the believability of the project.  

An example of wisely pulled punches: 
...our model does not have any role for business activity to affect TFP growth. We agree that TFP growth is not automatic, and much of it originates with activities related to innovation and entrepreneurship. Lacking a quantitative understanding of this process, we treat them as zero in this paper but do not deny the assertion that additional taxes and regulations would reduce the growth of TFP and not just its level.

Their bottom line was only 8.5% reduction in the level of GDP. It is entirely worth speculating whether Bidenomics would reduce the growth rate of GDP, driven by innovation. Any such effect would dwarf the level effect eventually. But "Lacking a quantitative understanding of this process," they say nothing. 

At another level, though I have to be a bit critical. As impactful writing, do not take this report as a model of rhetoric! It is nearly unreadable if you're not a well trained economist. They put the tables in the back where you can't see them! The report starts with taxes, which are eye glazing, not the more innovative analysis of regulations. 

**** 

The Letters 

The Nobel Prize Winners offer a short letter. Its core: 

we believe that Biden’s overall economic agenda will improve our nation’s health, investment, sustainability, resilience, employment opportunities, and fairness and be vastly superior to the counterproductive economic policies of Donald Trump.

Throughout the coronavirus crisis, Biden has recognized that science-based, public health solutions are critical not only to saving lives, but to any viable strategy to restore economic confidence, recovery, and jobs. Similarly, on issue after issue, Biden’s economic agenda will do far more than Donald Trump’s to increase the economic strength and well-being of our nation and its people. Simply put, Biden’s policies will result in economic growth that is faster, more robust, and more equitable.

Now, the Biden plan at its core consists of a sharp rise in tax rates, pushing many marginal rates well past 50%, a pledge to both spend and via unions and other regulations deliberately increase economic "wedges," and to deliberately use more expensive but less carbon-emitting energy. Such a plan might be defended as an important sacrifice of economic growth for greater equality, or for environmental benefits not quantified by GDP, or for social justice. Indeed "less growth but more equitable" might -- might -- be a defensible position. But just how does it produce "economic growth that is faster?" 

The signatories turn economics on its head. OK, they have Nobel Prizes, I don't -- maybe all the standard signs of the effects of economic policies are wrong. But how? The offer no analysis, no mechanism, no quantification, and not even citation to others who have done so. Just their "belief" and  authority as Nobel Prize winners. Or is it just their moral judgement and partisan feeling, no less sincere, but not any more special than yours and mine? 

The economists against Trump offer a personal list of charges, reminiscent (deliberately, I hope) of the charges against George III in the Declaration of Independence

His chaotic and ineffective approach to negotiation has damaged relations with trade partners...

His managerial incompetence has damaged the credibility and effectiveness of the public sector....

His administration’s public health response to COVID-19 was described by medical scientists as having turned "...a crisis into a tragedy" and as having underperformed relative to other democracies by "orders of magnitude."

Ah, dear party of science enthusiasts, just which democracies other than maybe New Zealand do you have in mind here? Europe is exploding with covid-19, and last time I looked Trump does not rule them  

He has consistently undermined the independence and credibility of our major health agencies,...

He has dramatically overemphasized the extent to which economic growth and public health are in conflict during a pandemic. In fact, many countries that have been more effective in their viral containment efforts have also performed better economically.

Umm. Sweden? 

His personal behavior during the COVID-19 outbreak...

He regularly spreads dangerous misinformation...

And so on. The best

He has a poorly-informed, zero-sum view of economics that engenders needless viciousness and cruelty.

Really? And your redistribution, unions, and various adjective-justice plans are not about zero-sum economics, but just growing the pie? 

OK, I get your outrage at the President's behavior and his twitter account. Lots of passionate partisans including Republican Never-Trumpers feel that way.  But you're writing as economists, trading on what's left of our profession's claim to scientific expertise, to objective analysis of policy and its outcomes. What reader, seeing your signature on this letter, would dream to believe that you can offer documented, reproducible policy analysis free of partisan bias? 

(Admittedly,   

He claimed to have the unique ability to generate growth (in real GDP) of between 4% and 6%, but never surpassed 2.9% in his first three years in office. Furthermore, analysts at Goldman Sachs and Moody’s Analytics have projected that Joe Biden’s economic plans, if implemented, would actually generate faster growth in both employment and real GDP.

So they do cite a study, though they don't give a link, and interestingly not one produced by academic economists. I have not read either, but the WSJ characterized the Moody's study in the nakedly Keynesian terms I did above.  The incremental economic analysis relative to partisan passion in this letter remains zero.  And in fact GDP did regularly come in above forecasts in the Trump years, and regularly under forecasts in the Obama years. But that fact may reflect more on how forecasts are made.)  

The economists for Trump do not do much better. This one came later, and at least I can sympathize with a desire to let the public know that the entire profession is not in one partisan camp, but really all this does is to let them know there are a few economists in the other partisan camp. 

We enthusiastically endorse President Trump's re-election efforts on the basis of his track record and continued economic policy agenda. 

Prior to 2020, the President's pro-growth economic policy agenda included regulatory reform and tax reform that among other changes reduced the federal corporate tax rate to 21%. We believe this meaningfully contributed to accelerate GDP growth  and historically low unemployment which fell the the lowest level in half a century. Wage growth also picked up during this period for the first time in decades and 2019 saw the largest single-year percentage gain in median household income on record (a 6.8% increase) to $68,700. These economic gains have been widely shared across-the-board for Americans. ..

At least this one had some economic facts, which are uncontestable. 

We believe that Vice President Joe Biden's proposals to add regulations and raise taxes amid an economic contraction, namely raise the federal corporate tax rate to 28% from 21%, will meaningfully hinder the economic recovery and slow the process of Americans getting back to work.

Yes but why? Whence this "belief" and who cares about your beliefs. At least they are not presuming, as the anti-Trumpers do, that our stature as economists gives us special insight to judge the moral character of politicians. But though I agree with the analysis (I explicitly offer no political endorsement, that's not my job and you should not care who I vote for), analysis needs to be analysis to add to public debate. 

Matt Yglesias tweeted recently

We need to some day [to] try to swim back to the idea that partisan politics is a contest about who will run the government and what will they do with that power, not just a referendum on diffuse cultural trends with tenuous connections to policy.

In public when using your professional status to claim expertise, you should be in that business. 

Update:

After a few days, I have reflected a bit, and at least come to a more nuanced view. After all, I express lots of political opinions -- see Understanding the Left.  I offer views of economic policy. That's basically the point of this blog. And I don't always footnote everything with detailed studies. That's putting it mildly. Most of what I offer is a rough sketch of whatever literature I can remember and my own views as I try to sort through what sounds right and what does not. 

So why am I so annoyed at the letters? Am I not a little guilty of some of the same sins? 

Now, I don't express my views of the moral character of politicians, nor do I endorse candidates in public. The letters did that.  I try hard to separate economic analysis from political judgement, and remember the rule that both parties are sinners. The letters let their partisan passions and judgements about Trump's character bleed in to economic judgements, that somehow sharp raises in taxes and regulations are going to increase economic growth. 

But everyone is entitled to their opinions and to spout off as they wish.  I spout off about plenty of other topics on which I have little special expertise. And I unashamedly work hard to preserve a free market view, more quickly accepting confirmatory facts and logic, and more cautiously to jump on the latest puzzle. 

What annoys me the most, I think, is the essential character of a joint public letter. It does not say, I Jane Q Citizen, think the current occupant is a jerk and should be thrown out. It says we, as economists, representing the economics profession, summarizing scientific knowledge proclaim the current occupant is a jerk and should be thrown out. 

After all, why make it joint, and not just send it from each one's twitter account, and label it personal opinion? 

Making it joint, and signing it, also signals the social coercion so abundant in other academic fields and infecting economics like covid-19 in a frat boy back from a rave. Hey young assistant professor, all the bien-pensant people in your profession -- all your tenure committee, letter writers, editors and Nobel prize winners -- are on the bandwagon here. 

The point of a joint letter is to try to demonstrate unanimous opinion, both outside and inside, not just we, who happen to be economists, think that Trump should be ousted, but all good-thinking economists should join us. 

For that reason I was a little kinder to the joint pro-Trump letter. At least it signals, in response, hey, it's ok to have a differing opinion. Sure, the economics profession may be 95% for Biden, but you are not totally alone, there is a group that feels as you do. 

I signed one joint letter a long time ago -- economists opposed to the TARP. I still think it was a good idea -- it revealed to the world that there were some economists opposed to TARP, and I think we did some good in slowing down a steamroller headed in the wrong direction. The joint letter opposing the Smoot-Hawley Tariff was a great joint letter, for all the reasons these are bad joint letters.  

Bottom line, to all of us: think long and hard before organizing and signing joint letters on political matters. 




27 comments:

  1. Kevin Hassett was one of the signatories to the Romney Economic White Paper that blamed the housing bubble and the Great Recession on the CRA so I'm not going down any rabbit hole that he has helped dig.

    ReplyDelete
    Replies
    1. "Kevin Hassett [signed] the Romney Economic White Paper that blamed the housing bubble and the Great Recession on the CRA"

      The CRA required the Fed and other federal banking regulators to encourage —encouragement by the fed is usually accompanied by strong suggestions, threats, and a club— financial institutions to meet the credit needs of low- and moderate-income (LMI) neighborhoods. It can be likened to a snowball at the top of a hill. It only takes a slight push to get it rolling downhill.

      That push came in 1992 when the 102nd congress weakened regulation of Fannie Mae and Freddie Mac. This allowed Freddie and Fannie to reduce liquidity from 10% to 2.5% with the intent of buying more mortgage loans; and as it evolved, higher risk (LMI) mortgage loans.

      As it rolled downhill, the snowball (CRA) picked up mass and was accelerated by the passage of the Gramm-Leach-Bliley Act in 1999 which repealed part of the Glass–Steagall Act of 1933. The repeal contributed to the proliferation of the complex and opaque financial instruments at the heart of the crisis.

      In defense of Mr. Hassett, he was right in blaming the CRA as the start of the housing bubble that initiated the Great Recession.

      Delete
    2. "But this is a general failing. Where is the estimate of how many tons of carbon the whole green new deal reduces, how much global temperatures decline, and how much GDP is thereby saved in 2100?"

      How many tons of FUTURE carbon is reduced. How much air carbon present today is "OK". If you think the carbon air density will not grow then I suppose the Biden plan should show a reduction.

      Delete
    3. Mr Blough - here is an article addressing the allegations about the CRA. There are others that make the same point.

      https://ritholtz.com/2016/06/no-cra-not-cause-financial-crisis/

      Delete
  2. This comment has been removed by the author.

    ReplyDelete
  3. -Well, the Nobelists are from the coasts.

    -The 1072 are the usual suspects plus the wannabees.

    -The Trumpistas are not inserito.

    No surprises there. This is fit for a project in social psychology, not economics.

    Pity it's come to this.

    ReplyDelete
  4. I gather that Silicon Valley, Wall Street, academia, the foreign policy-military establishment, major media, and Hollywood all support Biden for president. Particularly, enterprises with business exposure in China.

    My guess is the groups above expect a better result for them, under a Biden Administration.

    More and more "carve-outs" for this or that special interest group?

    Detente with an increasingly repressive Communist Party of China?

    PS. "He [Trump] has a poorly-informed, zero-sum view of economics that engenders needless viciousness and cruelty."

    If you believe that, then what words are used to describe President Xi?

    ReplyDelete
  5. How does one obtain a greater rate of growth in macro economics?
    Start with the basic equation: Y = C + I + G + X. Differentiate totally: dY = dC + dI + dG + dX. Assuming dG > |dC + dI + dX|, dY > 0. Multiply through by 1/Y to obtain dG/G•G/Y > |dC/C•C/Y + dI/I•I/Y + dX/X•X/Y|, and dY/Y > 0. Provided dG/G is sufficiently large, dY/Y can be greater under an incoming president than under his predecessor. This is the approach taken in the "Joe Biden Plan": government expenditures exceed private consumption and private investment retrenchments.

    Reading the "Joe Biden Plan", I am left wondering what role there is for the private sector in the plan--I can see very little for the private sector other than perhaps some minor role as a government 'poodle' in some of the schemes that the plan advocates shall 'incent' private resources to co-invest in (but, just how this shall be undertaken is nowhere disclosed).

    Given the list of names of those associated with the various elements of the "Joe Biden Plan", one can easily imagine that the process of 'incenting' private co-operation with an incoming administration's priorities might entail a certain amount of public shaming and 'behind the curtain' presidential arm-twisting of those corporations vulnerable to those sort of actions. Not that this will rise to the level that a totalist-oriented government might aspire to, but it will be at a level heretofore not seen except during war-time, perhaps. The hectoring won't be quite so public, but it will be at least as desperate as the outgoing administration's was.

    ReplyDelete
  6. I saw the same Hoover report and had two immediate thoughts

    1) This is a nice bit of quantitative economic analysis of the Biden proposals and is more rigorous then what one normally sees

    2) It is rather selective both in terms of policies considered and models employed to estimate impacts

    3) At the end of the day, each side has their own numbers and the amount of uncertainty and imprecision in economic analysis is so large as to make such analyses largely if not entirely irrelevant.

    Economics is not like physics. Our ability to predict future outcomes based on policy parameters and our understanding of the economy is just too limited to be of that much use in informing such decisions. For (nearly) any estimate side A provides, side B can come up with quantitative rationales that are not crazy that would show "Not what A said".

    ReplyDelete
    Replies
    1. Economics is not like physics.

      Economics is like physics in one way, the only way that matters: It tells us what cannot happen.

      Delete
  7. Returning to coronavirus policy, the following John Cochrane quote was just reported by CNN:

    "Wise public policy has to balance a bit of disease spread against immense economic [and] social costs," said Cochrane, an economist. "I think that underlying insight drives much of [Scott Atlas'] particular recommendations."

    As far as I can tell, Scott Atlas himself has never said such a thing, at least not publicly.

    Nevertheless, I've not seen John Cochrane come this close to acknowledging the powerful role of ideological priors in public policy opinionating--including his own. The more sunlight shown on this the better.

    Finally, Cochrane's quote may be read as implying that the rest of the medical policy establishment is remiss for totally disregarding economic tradeoffs. On what evidence does he base this claim?

    ReplyDelete
  8. John you are an academic who as spent time in many institutions. If the majority understands that the simple Keynesian story is wrong, why are they continuing to peddle it? Its doing a lot of harm that this rhetoric continues generation after generation. Its totally understandable that public choice plagues the politicians into peddling this, but what gain comes from professional economists from doing it? Can anyone haphazard a guess?

    ReplyDelete
  9. Ah, redistribution via taxation - reminds me of Pareto efficiencies and improvements. No one likes having their cash pulled from their accounts, regardless if they're an employee or a business owner. The two pillars of economic thought with regards to individuals and firms automatically sets up a condition for rivalry between the two.

    And yes, individuals bear the brunt of taxes, not so much for corporations. How can this be improved? I'm not sure. Taxes go up, firms threaten to leave and leave the local populations destitute. This still goes on. And yet, our government seems powerless to find ways to improve this condition.

    Incentives definitely matter, yes. But, I think as a society, we have to change our preferences to have a society and government that functions well, especially the institutions people rely on. Trust matters a great deal.

    ReplyDelete
  10. Even if Trump's economic agenda were to be better for the economy and for growth of that economy than that of Biden's (and I doubt that, and not sure that Trump has an overall agenda rather than just whatever strikes his fancy at the moment), Trump's disruption to the movement to ever freer international trade will most surely damage world economic growth) adopting Trump's instincts is not worth the ruin it has taken to United States' alliances and foreign policy and to the checks and balances of our system of government. But I commend you for an attempt to offer professional unbiased analysis of Biden's plans (or was it just a rant against the lack of analysis from the Nobelists?), but remember the Biden plan is as much a statement of campaign strategy as it is of economic strategy. But your point is well taken that the economists who offered pro Biden analysis also seemed to be in campaign mode more than in professional economist mode. I still hope to high heaven that Biden wins so we can be rid of Trump who is so unfit to be President of the United States.

    ReplyDelete
    Replies
    1. Who has benefited from free international trade? Wall Street, politicians, elites and Chinese communist party. If allow this going for another five or ten years, the western civilization will be under the control of the communist government. Look at Hong Kong.

      Delete
  11. Speaking of economists and models, whatever happened to Kevin Hassett's cubic model of Covid deaths? Should the utter failure of this model temper our belief in the quality of his analysis, no matter how numeric or precise it purports to be?

    ReplyDelete
    Replies
    1. All analysis relies on background information regarding the dynamics at play. The difference between a quantitative model and the off the cuff comment of your uncle is that the quantitative model makes EXPLICIT assumptions and relies on EXPLICIT estimation or calibration of model parameters.

      I could not even to save my life determine what plagues the imagination of, say, Paul Krugman. However, when Paul Krugman publishes a paper, he is forced to be naked: all the assumptions are there, all the math is there, all the implementation details and results are there (or, can be found on demand). This means I can have a meaningful dispute with him over specific segments of his analysis in a way I could not otherwise.

      Quantitative analysis is not about an obsession over precision, but about being able to make meaningful comparisons in a transparent manner. Regardless of how I proceed (structural vector autoregressions, DSGE models or partial equilibrium models), I would not be comfortable making a strong statement if my results went out the window as soon as mild changes were made to the procedures I implemented. We CAN get conclusions that rest on balancing quantity on a knife's edge, but nobody trusts them. Likewise, we never talk about a fiscal multiplier of 0.89596857636 because that can only come out of a very specific procedure. But, if you read a good load of papers with different models, methods and datasets, you get a sense of what might be plausible.

      Delete
  12. Baseline seems to be what if the economic policy isn't implemented, when actually it's Biden's or Trump's. Where are the Trump policy costs? The trade war promotes more growth than the tax raises? Trump's tax cuts don't lead to long run problems? Kudos for the academic approach regarding the analysis, but the implications are all wrong as are the politics.

    ReplyDelete
  13. I agree a great piece, I missed it when I put out my piece on the strategies for setting up for a Biden win. I think the piece is far superior to the others I've seen such as from the Tax Foundation, Tax Policy Center, Moody’s, and Penn-Wharton Budget Model. The Hoover framework results in implications that get closer to the likely outcomes because it gets closer to how people and institutions make decisions-my Chicago biases shows. Over and above the macro implications of the model. The part I found unique was their analysis of Biden’s health care and climate change plans. However, I constrain my comments to their Macro conclusions.
    The Hoover piece results suggest the Biden plan will have a much greater impact on the economy on an absolute basis than any other piece I’ve read. They find that the aggregate macro impact of a decline in consumption of $1.5 trillion, median income down by $6,000, and GDP down by $2.6 trillion by 2030. However, the aggregate numbers understate the impact on an individual basis and could miss some of the convexities from the income redistribution aspect of his plan.
    Income redistribution is at the core of the Biden plan with as much as 75% of the tax policy changes--and likely--regulatory policy impact will hit the top 1% of incomes, and 100% will be absorbed by the top 20% of incomes. So, the top 20% of households, which represent 40% of consumption, will take the entire hit of the $1.5 trillion of aggregate consumption decline. Further, the income of the top 20% take on most if of the $6,000 decline in median income.
    One of their conclusions is that equilibrium capital levels will fall because the marginal tax rates on capital investment will go up. This likely understates the fall in equilibrium capital levels because it does not reflect the higher capital cost from the distributional aspects of the Biden tax policy. 100% of domestic net savings in the economy come from the top 20% of incomes. Accordingly, net savings in the economy will fall pushing up the cost of capital as supply of capital dries ups and driving down the level of equilibrium capital, and lower GDP even further. The potential crowding out from rising deficits and yet another stimulus package in 2021 could push cost of capital even higher. However, any good neoclassical model incorporates the Ricardian effect-which has a very strong empirical support. The effect would push households to save more and offset the higher deficits, which would push down the consumption of the top 20% of incomes even greater.

    ReplyDelete
  14. There's lots to criticize in the 'economists against Trump' letter, but their claim about Trump's zero-sum view of economics is not one of them. That is indisputably true, and pointing out that the other side may hold a similar daft view is irrelevant — two wrongs don't make a right.

    ReplyDelete
  15. Neither plan reduces the deficit....neither plan does anything productive with entitlement spending...neither plan is a serious effort to put freedom of choice in the hands of Americans, or what they might characterize as the prolateriat....Which one is "less bad"? Trump's plan is "less bad" than Biden's if expected GDP growth is higher....but neither are good.

    ReplyDelete
  16. As a PhD student in the field, I am obviously partial to the quantitative analysis performed in applied research. Depending on the topic and the data, methods can be more or less involved in the theoretical restrictions they impose, though I suppose that for tax policy at a macro level, DSGE models really are the only sane option.

    I personally have grown extremely tired of the utter lack of common sense in public debate surrounding the Trump administration. People going as far as supporting large corporations engaging in censorship and getting knee deep into swaying public opinion right into the election is not reasonable.

    As you correctly point out, the main benefit of economic modelling is their transparency: once you explain them well and clearly, anyone can dispute any part of the analysis -- and, more often than not, we can even directly change some parts of the analysis to see if incorporating the critique changes the results. It's wonderful -- and a hell of a lot more sophisticated than "Orange Man Bad."

    ReplyDelete
  17. "The economists against Trump offer a personal list of charges, reminiscent (deliberately, I hope) of the charges against George III in the Declaration of Independence"

    Each point you reference from the letter makes specific claims and two of them cite external evidence.

    Consider the first point, which you partially quote and satirize, rather than engaging:

    "His chaotic and ineffective approach to negotiation has damaged relations with trade partners, interrupted supply chains, degraded international norms, and harmed American farmers -- all without achieving his stated goal of reducing the trade deficit. Even his landmark trade deal, the USMCA, is not projected to have a meaningful impact on either GDP or the trade deficit."

    Do you think his approach to trade has been orderly and effective? Do you think it hasn't damaged relations with Mexico and other trade partners? Do you think it had a substantial impact on the trade deficit, contrary to the official statistics? Do you disagree with the IMF's report on the USMCA, which finds that it is unlikely to have a substantive impact on GDP or deficits?

    What, exactly, is wrong with trying to convey this information to the general public?

    "Ah, dear party of science enthusiasts, just which democracies other than maybe New Zealand do you have in mind here? Europe is exploding with covid-19, and last time I looked Trump does not rule them"

    The letter references an external argument from the NEJM, but go ahead and pretend it says something else.

    "Umm. Sweden?"

    Compare Sweden to its natural reference group: Norway, Denmark, and Finland.

    "Really? And your redistribution, unions, and various adjective-justice plans are not about zero-sum economics, but just growing the pie?"

    The letter says nothing at all about redistribution, unions, or "adjective-justice" plans. You actually just made this up.

    "So they do cite a study, though they don't give a link, and interestingly not one produced by academic economists. I have not read either, but the WSJ characterized the Moody's study in the nakedly Keynesian terms I did above."

    You first imply that there is something wrong with using work by private sector economists. You then attempt to dismiss work done by economists using a newspaper article.

    "But you're writing as economists, trading on what's left of our profession's claim to scientific expertise, to objective analysis of policy and its outcomes. What reader, seeing your signature on this letter, would dream to believe that you can offer documented, reproducible policy analysis free of partisan bias?"

    The 2016 letter made a long series of claims that mostly turned out to be true: https://sites.google.com/site/economistsagainsttrump/. Actually being right about things might improve our credibility with the general public more than creating increasingly sophisticated models that still can't beat a VAR.

    ReplyDelete
  18. Truly wonderful review of the Hoover analysis.

    Specially because it lets us know that you believe that government intervention in R&D can't increase TFP.

    Exactly the opposite is true: Government intervention in technology and R&D has historically been a key driver of growth. Even the internet with which we communicate is a testament for this.

    I therefore think that the Hoover report, which admits that the economic "damage" aren't catastrophical, coupled with the possibly increased TFP from more R&D and the benefits from action against climate change leads us to be very optimistic about the future Biden presidency.

    ReplyDelete
  19. John, while I disagree vehemently with a lot of your views, I hold you in high esteem as a fair and decent guy (having heard this also from several of your students). However, the premise of this post is misleading, and that's putting it mildly. I have worked in a number of policy institutions and ALL OF THEM have a model(s) to do just these types of analyses, basically ALL OF THEM having the neoclassical growth model as a benchmark. I'm actually not too much in favor of that as these large scale models are hard to follow, which is why all said institutions have semi-structural modules and complement these with smaller models. But NO ONE is doing simple IS/LM. As for reproducibility, the European Commission's Quest model is publicly available.

    Perhaps you had something more specific in mind that did not come through in the post, but as I read it, the post claims that everyone uses IS/LM and these guys are the real deal because they do something that's never been done before. And that's just not true. I'll be happy to be proven wrong.

    ReplyDelete
  20. John,

    "Increase the corporate tax rate to 28%"

    Have you figured out yet that the ONLY way to get "permanent" tax breaks is to sell them through the Treasury department rather playing Peanuts football with them?

    Lucy the Republican politician sets the tax cut football down for Charlie Brown to kick, then Lucy the Democrat pulls the football away.

    The same thing happened to a lot of the Reagan tax cuts, the same thing happened to a lot of the Bush Jr. tax cuts.

    The definition of insanity - doing the same thing over and over and expecting a different result.

    This amateur economist does not consider Trump to be a jerk, he considers the entire Republican Party to be clinically insane.

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.