President Trump’s tax reform vs. his balanced budget

President Trump has a $2 trillion hole in his fiscal policy proposals. His numbers don’t add up. This creates a conflict between two of his fiscal policy goals: tax reform and balancing the budget.

Let’s look at three elements of President Trump’s economic policy and how they interact:

  1. Last month he proposed tax reform with most of the key numbers left blank.
  2. Today he proposed a budget that claims to reach balance in 2027 (year 10).
  3. His budget assumes his economic policies would increase economic growth by a lot, to 3 percent per year.

In addition to proposing a budget that purports to balance in year 10, today Trump Budget Director Mick Mulvaney told us one key new fact about tax reform: the Administration now assumes tax reform will be debt neutral. Director Mulvaney used this to explain why the President’s top fiscal priority, his tax reform proposal, which would involve trillions of dollars of changes to tax policies, was omitted from the President’s budget. This omission is, to say the least, odd.

There is significant public debate about whether Team Trump’s aggressive growth assumption is reasonable given the policies he has proposed. For now let’s set aside this critical question and pretend it’s reasonable. Let us assume President Trump’s economic advisors are right, that his policies would result in 3% real growth per year, and that this faster growth would benefit the budget. Let us further assume their estimate of the [budget] Effect of economic feedback is correct. You can see it in today’s budget proposal (Table S-2, near the bottom of page 26, which is page 32 of the PDF). President Trump’s advisors assume this faster economic growth will reduce the budget deficit by $496 billion in 2027, their target year for balancing the budget.

The President’s balanced budget claim depends on this $496 billion effect of economic feedback in year 2027. They assume almost $500 billion of government spending bills in 2027 will be paid from additional cash inflows that result from higher government revenues resulting from faster economic growth, rather than from cash borrowed from financial markets. Faster growth —> higher government revenues —> less need for government borrowing to pay spending bills —> lower deficits and debt & budget balance in year 10.

This $496 billion is a really big number for a single year. For comparison, it is almost twice as large as the $251 billion the president proposes to cut non-defense discretionary spending in that year. It is three times as large as the $165 billion the budget proposes to save in Medicaid in that same year.

On that same line of Table S-2 you can see Team Trump assumes economic growth means the federal government will need to borrow $2 trillion less over the next ten years. That equals 6.6% of GDP in 2027, an enormous amount. When Director Mulvaney says President Trump’s budget would reduce debt/GDP from 77% this year to 60% in 2027, about a third of that reduction is from this single assumption.

So far, so good. Items (2) and (3) work together: the balanced budget promise and the positive budget effect of the 3% growth assumption, which for now we are stipulating is valid. The problem is fitting debt-neutral tax reform into this puzzle as well.

We don’t know how much the tax reform proposal would cut taxes because in April President Trump did not provide sufficient detail to estimate it. The President’s campaign proposal was roughly a $6 trillion gross tax cut. Let’s make a wild guess and assume his new proposal is smaller, a $5 trillion gross tax cut. The concept that follows is what matters, not the actual gross number.

If your tax proposal, which you left out of your budget proposal, is debt neutral, then you need to have the same amount of new revenues to fully offset the revenue lost to the government from your proposed gross tax cut. In our example you’d need to have $5 trillion in new revenues over ten years to combine with $5 trillion of gross tax cuts to result in a debt-neutral package. In theory, this offsetting revenue can result either from proposed tax increases or from the higher revenue that results from economic growth, or from a combination of the two. You’d also need to match your revenue loss and revenue gain in 2027 so that your proposal doesn’t affect balance in that year.

In our example, if you combined $5 trillion of gross tax cuts with $3 trillion of tax increases, your tax reform package would be a net $2 trillion debt increase over ten years. If, however, your tax cuts would also result in faster economic growth, and if you think that economic growth would result in an additional $2 trillion of government revenues, then your tax package in total would be fully offset and debt neutral. This dynamic scoring of tax reform would make it significantly easier to enact debt-neutral tax reform, because you would need to add only $3 trillion of painful tax increase policies to a package that includes $5 trillion of gross tax cuts that people and businesses like and support.

But you cannot have it both ways. If you try, you are double counting. Either the $2 trillion of added cash inflows resulting from faster economic growth can pay for more government spending and reduce the need for government to borrow, or that $2 trillion can replace the cash lost to the government from cutting taxes and reduce the size of painful tax increases you need to propose. Arithmetic forces you to choose one goal or the other.

Last month Secretary Mnuchin counted the (then unspecified) positive budget effects of economic growth to help offset their tax reform package. Today Director Mulvaney counts those $2 trillion of extra revenues to reduce government borrowing and achieve a balanced budget. Logic requires they choose one or the other, but today they chose both and Director Mulvaney said that choice was deliberate. There will be only one $2 trillion stream of cash (if you even believe it’s that large). By claiming they can do two things with each dollar of cash they have left a $2 trillion hole, either in the Trump balanced budget proposal or in the Trump debt-neutral tax reform proposal.

If they’re going to use growth effects to help balance the budget as proposed today, then Secretary Mnuchin either needs to convince President Trump to support $2 trillion of additional tax increases to keep tax reform debt neutral, or they need to support significantly smaller gross tax cuts. Secretary Mnuchin has so far opposed the two biggest tax increases needed for a big debt-neutral tax reform: a border adjustment tax and eliminating the business deduction for interest expenses (which, coincidentally, would together raise about $2 T of revenues over 10 years). If they want to use dynamic scoring to make tax reform easier to enact, then President Trump and Director Mulvaney do not have a balanced budget proposal until they find almost $500 B of additional deficit reduction in 2027.

You can’t have it both ways, and $2 trillion is a big hole to fill.

Tuesday, 23 May 2017|

re: The President’s debt tweet

For the classes I teach at Stanford’s Graduate School of Business I make my students write policy memos to a friend or family member as if that person was a Member of Congress. I have done the same here. These memos are similar in style to those I used to write for President George W. Bush and Senate Majority Leader Trent Lott. Here’s a pdf version.

25 February 2017




You asked whether you should echo or retweet President Trump’s tweet about declining debt.

In a word, no.

It appears the president was repeating something Herman Cain said this morning on Fox & Friends Weekend. We know the president watches this show and his tweet appeared shortly after Mr. Cain was on-air.

The numbers are technically correct.

  • Debt held by the public declined $19.6 B from January 20, 2017 to February 23, 2017, the most recent day for which data is available.
  • In 2009 the same measure increased $222.6 B (more than the “$200 billion” the president cited) over the same timeframe.

But government cash flows are lumpy, leading to big daily fluctuations in government debt.

  • Had the president / Mr. Cain ended his timeframe one day earlier this tweet would have been invalid and debt would have increased (by just $1 B) in “the first month.”
  • This is why analysts look at debt on an annual basis rather than daily/weekly/monthly.

Neither president affected government borrowing in his first month.

  • Government borrowing in January and February is the byproduct of spending and tax policies set by Congress the year before. President Obama signed the fiscal stimulus law on February 17, 2009, but it took months before that began to change government cash flows and borrowing requirements. President Trump has so far not measurably affected fiscal policy in general or government borrowing in specific.
  • It’s unfair to assign any responsibility for borrowing in the first month to either president.

The big difference between early 2009 and now is the health of the economy.

  • GDP was plummeting when President Obama took office. Tax revenues were down, automatic stabilizer payments (e.g., unemployment insurance and safety net spending) were up, and funds were being spent from the Troubled Asset Relief Program (TARP). In early 2009 government was borrowing a lot because the economy was weak, not because of President Obama’s policies.
  • In contrast, the U.S. economy is now growing. The smaller borrowing requirement for this month is mostly a result of this economic difference, and may also in part be simply an artifact of choosing such a short timeframe for comparison.


Because of his unique communications advantages, President Trump may be able to get away with making an argument with such a weak foundation. You cannot, and you should not place yourself in the position of having to address the intellectual weaknesses described above.

More concerning, this tweet shows the president continues to rely on TV rather than his advisors for numbers and policy substance. Until his staff figure out a way to ensure he doesn’t make such easily rebutted claims, you should not echo the president’s economic arguments or claims without first verifying both their accuracy and substantive merit. This is particularly true of his early morning and late night tweets, when he’s probably in the residence and away from his staff. This unfortunate situation will persist as long as President Trump continues to take his numbers and policy arguments from TV pundits rather than from Mr. Cohn, Director Mulvaney, and Secretary Mnuchin.


Saturday, 25 February 2017|

Jobs, jobs, jobs. And also…

Last Tuesday President-elect Trump said, “My Administration will be focused on three very important words: jobs, jobs, jobs.” Mr. Trump emphasizes geography: American policies should encourage economic growth “right here in America.” Nothing wrong with that.

He follows a hallowed tradition of politicians emphasizing jobs and talking about the number of people employed as if it were the only measure of economic well-being. This is effective political communications about economics and quite common.

Any time a policymaker does this he or she is oversimplifying. Employment levels, that is, jobs, are an important metric of economic health. Maximizing the number of people working is a laudable goal for policy. But jobs are not the only thing we should care about. Economically we are more than just workers.

  • We care not just about how many people are working but also about how much we are earning. We are wage-earners.
  • We care not just about wages but also about non-wage benefits like employer-paid health insurance premiums and employer contributions to retirement savings. Wages plus benefits equals total compensation.
  • We care about the prices of the goods and services we buy. Those prices can be affected by the extent of domestic and international market competition and also by inflation. We are consumers.
  • We can only spend what the government doesn’t take from us, so we care about the taxes we pay. We are taxpayers.
  • On the flip side, we may receive cash or other transfer payments from the government. Some of us are recipients of government benefits.
  • And since many of us own financial assets we care about economic growth and its effects on financial returns. We are investors.

Politicians talk about economic policy as if having a job is the only thing that matters to you. While being employed is critical to your economic status, you are more than just a worker.

When thinking about economic policy, each of us is a worker and a wage-earner, a consumer, a taxpayer, and often an investor and a recipient of government benefits.

OK, but so what? Does it really matter if politicians oversimplify and talk only about jobs, rather than about jobs and wages and compensation and prices and taxes and benefits and market returns?

It matters if political rhetoric drives policy to prioritize one element over the others. If we ignore a policy’s effects on consumers and taxpayers and measure only its impact on jobs, then we are getting an incomplete view of that policy. If we measure a policy’s success only by counting how many more jobs will result without considering the effects on wages and prices and other measure that matter, then we’re going to choose bad policies.

I am not arguing we should ignore the employment effects of a policy; I am not arguing that jobs don’t matter. Quite the opposite. We should start by looking at the effects on jobs and then keep going, analyzing the effects on wages, non-wage benefits, prices, taxes, transfer payments and financial returns. We need to measure, estimate, and understand all the economic effects of a proposed policy change, not just the one effect that is easiest to communicate.

Example 1: Buy American provisions, aka domestic content restrictions, require the government to buy certain things only from American producers. This helps steelworkers while harming taxpayers and those who use the government’s services (a fixed number of tax dollars buying more expensive steel will buy fewer rail cars and buses). We should evaluate these policies not just on the American jobs they create or protect, but also on the Americans’ taxes they increase and the services they reduce for other Americans.

Example 2: Erecting trade barriers with China and other sources of inexpensive imported goods would help American workers in the protected industries, at least in the short run. It would also raise prices for those products. We should consider the effects on both American manufacturing workers and on American Walmart shoppers. If we ignore the latter, much larger group, we’re not making good decisions.

Example 3: A large deficit-increasing increase in government spending or cut in taxes next year (aka fiscal stimulus) could, by itself, increase employment and wages. It could also lead to higher inflation and harm consumers. Or the Fed could react to the fiscal stimulus by raising interest rates more rapidly. This would slow growth in interest-sensitive components of our economy, at least partially offsetting the effects of fiscal stimulus. Higher interest rates would also affect financial returns, the ability of American farms and firms to export goods, and the prices of things we import. Policymakers need to look at the complete set of impacts of a proposed fiscal stimulus, not just at the gross number of jobs someone thinks it will create.

Jobs are important and Mr. Trump deserves credit for signaling early that he will prioritize faster economic growth. We should not single him out for focusing his rhetoric solely on jobs, as most other politicians do the same. At the same time, each of us is more than just a worker, and the number of jobs is not the only measure of a good policy. When we forget this we risk doing more harm than good.

Monday, 19 December 2016|

How much advice does the president need? A simple example

How much advice does the president need? Should the president rely on just one person for advice on a straightforward policy question, or delegate a decision to a single person?

Challenge #1: Any policy problem difficult enough to make it to the president’s desk is usually multidimensional.

Challenge #2: The work of government is often highly interconnected. What one department does affects another.

Challenge #3: Once you get past the analysis, most presidential decisions involve tradeoffs among competing values and interests. I believe those judgment calls should be made by the person selected by voters.

Let’s construct an imaginary example of what appears to be a straightforward, even simple, policy question, and then use that to understand how policy advice to a president typically works.

Suppose House & Senate Republicans are moving a bill to increase government infrastructure spending by $200 B over the next decade, fully offset by cuts to other government spending. Let’s pretend Senator Ron Wyden, senior Democrat on the Senate Finance Committee, proposes doubling that new spending to $400 B over ten years with the increment to be offset by a roughly ten cent per gallon tax increase on gasoline and diesel fuel. Suppose Senate Minority Leader Chuck Schumer has told White House Chief of Staff Reince Priebus that he wants to talk to President Trump about this proposal. Finally, suppose President Trump decides he wants advice on how to respond to Senator Schumer.

Thus the question is: How should President Trump respond to Congressional Democrats’ proposal to double the proposed government infrastructure spending increase to $400 B, offset by a tax increase of roughly ten cents per gallon on gas and diesel fuel?

Pretty straightforward, right? Let’s unpack the question the way a White House policy council staff would.

Policy dimensions

  • Numbers: Are Senator Wyden’s numbers right? Will a ten cent per gallon fuel tax increase raise $200 B over ten years? A tax increase will increase the gross pump price, which will reduce demand for fuel, partially offsetting the initial price increase. What is the net effect on the pump price from a ten cent per gallon gross increase?
  • More infrastructure: How much should the president value an additional $200 B of infrastructure spending? How much more economic growth and how much better of a quality of life will an additional $200 B buy? What does it mean for people and stuff traveling by air, rail, car, truck, bus, and ship?
  • Economic effects of tax increase: What are the economic effects of tax and price increases of that size? Will they slow growth and by how much? What are the distributional impacts: regional geographic distribution, urban vs. rural, income distribution? What are the sectoral impacts: trucking, agriculture, Uber/Lyft/taxi, the transport component of the cost of consumer goods?
  • Environmental: More expensive fuel —> less driving —> less pollution and fewer greenhouse gas emissions. How big are these effects and how do they compare with other policy tools like fuel economy requirements and limits on power plant and industrial emissions?
  • Energy supply and demand; national security: How much will domestic fuel demand decline? How will that affect domestic oil and gasoline suppliers? How will it affect U.S. oil and fuel imports and exports? Will those changes affect our relationships with foreign oil suppliers like Venezuela and the Middle East?
  • Other increments: What if we did half what Senator Schumer is offering, +$100 B for a 5 cent increase? What if we doubled Schumer’s proposal to $400 B / 20 cents? What if we did Schumer’s +$200 B paid for half by gas tax and half by further spending cuts, or all by spending cuts?

Legislative, political, and communications dimensions

  • Communications: Fuel tax increases are extremely unpopular and communicating support for them is hard. What is the best communications strategy? What exactly will the president and his Administration say?
  • The politics are multi-dimensional:
    • left-right: The more
      [economically] conservative you are, the more you tend to oppose any tax increases, including these.
    • urban-rural: City dwellers are affected less than rural folk. If you drive your F-150 forty miles to and from work each day, you care a lot about fuel taxes.
    • infrastructure spending: Almost every Member of Congress wants to spend more money on infrastructure. Most will suck up a lot of other political pain if it means fixing a bridge in their district or State.
    • campaign fear: The campaign attack ads against an incumbent who voted for a gas tax increase write themselves. I wonder if I can get an ad on that little TV built into the gas pump?
    • political cover: Bipartisanship and/or presidential support might partially mitigate someone’s re-election risk.
    • States and localities: When the Feds raise these taxes it makes it harder for Governors and Mayors to do the same, triggering political pushback from those officials.
  • Legislative:
    • votes: How many Republican votes do we lose in the House and Senate if we support including a gas tax increase? How many Democrats do we pick up? Would we risk a Senate filibuster on the right? What would the votes look like if we counter-offered a different pay-for?
    • leadership support: Would Ryan and McConnell go along with it if the president insisted? They control the legislative process and without them we’re sunk.
    • other effects on the bill: If we lose Republican votes do we have to make other policy sacrifices because we now rely on Democrats to pass the bill? Will Congressional Democrats then demand further changes to the type or location of spending, or to union-related or environmental provisions?
    • legislative linkage: How would splitting Congressional Republicans and antagonizing their leadership weaken our ability to get them (the majority party) to help us with other parts of our legislative agenda? Would an alliance with Schumer on this bill provide other legislative benefits?

Someone needs to be able to educate the president about all these aspects of this decision as well as to answer these questions as needed. Let’s look at the president’s most senior advisors: Cabinet Secretaries and the most senior White House staff. Which of them have information, expertise, and/or formal responsibility for these questions and should be given an opportunity to advise the president on this decision?

Two Cabinet-level officials are particularly important:

  • Secretary of the Treasury because it’s taxes; and
  • Secretary of Transportation because of the infrastructure spending.

Six additional Cabinet-level officials can make strong cases that they should be part of the discussion:

  • EPA Administrator because of the environmental impacts;
  • Secretaries of Interior and Energy because of the effects on oil supply (Interior) and demand and import/export (Energy);
  • Secretaries of Commerce and Agriculture because of the sectoral impacts;
  • Secretary of Labor if a legislative alliance with Democrats necessitates changes in labor policy.

So far we’re between two and six Cabinet level officials who probably should be part of this decision. Now let’s look inside the White House (technically the Executive Office of the President, which includes the White House as well as other offices). We’ll start with the in-house policy experts.

  • The President needs to have the Director of the Office of Management and Budget who manages both the spending and tax sides of the ledger. $200 B is a lot of money.
  • He’ll need his chief economist (Council of Economic Advisers) to answer the economic questions. In theory maybe SecTreas or SecCommerce could do that, but while those two have economists working for them, you’d like to have an economist principal in the room. That’s the CEA Chair.
  • If you include EPA you’ll also include the Chair of the White House Council on Environmental Quality (CEQ).

Then you need the non-policy advisors in the White House:

  • the head of White House Legislative Affairs to address all the legislative complexities described above;
  • the president’s White House political advisor to advise on how this affects politics, including popularity, partisanship, and interest group support and opposition; and
  • the communications director and the press secretary to advise on how this fits (or doesn’t) in the president’s overall message and on how to talk about it publicly.

We’ve added another seven White House advisors so far. Just a few more:

  • the Vice President;
  • the White House Chief of Staff;
  • the Deputy Chief of Staff for Policy; and finally,
  • the Director of the National Economic Council, who corrals all these people, coordinates the process, produces the paper and runs the meetings.

Phew! That means the skinny version of this decision, with only two cabinet secretaries, has twelve senior people who merit advising the president on this simple question. The fuller version has up to nineteen advisors, each of whom has a strong argument for participating in the discussion leading up to this presidential decision. In practice nineteen is too many; you’d probably end up with around 15. And the president might take input from a dozen or more senior advisors, then have a follow-on smaller discussion with just a few of them. Even so, a lot of people need to advise the key decision-maker, even if they’re not all in the room when he makes the final call.

Why do you typically include so many people in providing input into a presidential decision?

  1. These advisors have information and expertise—Each question listed above requires analysis and research. The experts who do this work report to these 12-19 principals throughout the Cabinet and White House. These principals also have personal expertise, experience, and judgment (if you hired right). And they are each in close and frequent contact with relevant constituencies (Congress, truckers, farmers, drivers, labor unions, the press, farmers…) and can therefore add useful context and texture to a policy discussion.
  2. They have formal jurisdiction—Part of their job is to advise the president on policy issues within their portfolios.
  3. Cabinet management, morale, and policy implementation—These senior advisors are successful people who were told they’d have an impact on policy. They’re working hard on the president’s behalf every day. Many have not-small egos. If you exclude them from the big decisions you’ll have to spend more time and energy managing them, and you’ll have a tougher time getting their portions of the bureaucracy to faithfully implement the president’s decision.
  4. Cabinet effectiveness — A Cabinet secretary who is excluded from providing input directly to the president quickly loses the ability to be effective with others: Congress, the press, and interest groups. If you don’t have the President’s ear people figure that out quickly and work around you.
  5. Reduce insularity — Mixing up and expanding the personnel involved reduces the chance the president and his closest advisors will get trapped in a small groupthink bubble.

Could President Trump delegate a policy decision like this to one person, either a Cabinet Secretary or someone in the White House? Sure, but which one? Either you delegate to one person who likely prioritizes their part of the problem, leading to an unbalanced decision that doesn’t account for all of the President’s interests. Or you simply push the advise problem down a level: if you delegate to the White House Chief of Staff, or the Vice President, or the DCOS or NEC Director, then that person needs to get advice from these other dozen or more senior advisors. You haven’t solved the advice management challenge, you’ve simply relocated it.

All of the above was to advise the president on an apparently easy question. Imagine how complex it gets when you’re dealing with something hard. Can the president rely on one person for policy advice? Of course. A president can make decisions any way he wants. If he wants to make the judgment calls, and if he wants to make a well-informed decision, he needs information and counsel that represents a wide range of experience, expertise, and viewpoints.

Saturday, 17 December 2016|

What is a White House policy council?

What is a White House policy council and what does it do?

  • The President needs someone physically close to him whom he trusts to answer policy questions as they arise.
  • When he has to make a policy decision someone has to get him all the information he needs to make a good decision, and he needs someone to sift through and mediate the oft-conflicting views of his advisors. This is an honest broker role.
  • He may need someone with policy expertise to advise him on such decisions, someone who can see the big picture rather than just one part of it. This is an advisor role.
  • When he makes a policy decision that spans Cabinet agencies he needs someone to ensure the different parts of his government are coordinated, implementing his policy decision rather than their own preferences.
  • When problems crop up he needs someone to make sure they get solved, especially when the problems and/or solutions cross jurisdictional lines within the executive branch.
  • Finally, he needs policy experts to work with his other advisors to explain and sell his policies to Congress, the public, the press, and the world.

The staffs of the White House policy councils do all these things for the president.

President Trump will inherit from President Obama three White House policy councils:

  1. the National Security Council (NSC);
  2. the National Economic Council (NEC); and
  3. the Domestic Policy Council (DPC).

Each council has two components: the council itself and the policy council staff. There is a National Economic Council, which formally consists of eighteen people: the president, vice president, 12 cabinet-level officials and four senior White House staff. Then there’s an NEC staff within the White House, comprised of 15-25 people. Usually when people refer to the NEC they actually mean the NEC staff. The most senior of these has the title of Assistant to the President for Economic Policy and Director of the National Economic Council. President-elect Trump has named Gary Cohn to that position. Mike Flynn will run the NSC with the title National Security Advisor, and they still need someone to run DPC.

I know most about the NEC so I’ll use that for illustration. DPC is a parallel to NEC. NSC, the granddaddy of the three, is more than 15X the staff size (~350) and runs a bit differently, but the basic approach is quite similar.

The hypothetical examples in the first three bullets here are drawn from today’s Wall Street Journal.

  • Advice: President Trump continues: “Should we be doing anything about either of these? Should I tweet something about Facebook or Yahoo, either to praise either firm or to scold them? Do we need to do something from a policy standpoint?”
  • Policy development & decision support: More: “I see the Journal editorial page thinks we should fast-track approvals for liquified natural gas exports. That sounds like a good idea, as long as we don’t shortchange Americans. Get whoever we need together and present me with some options. Also, now that Harry Reid’s gone I think we should get Yucca Mountain restarted. Figure it out.”
  • Implementation: Imagine President Trump has decided to implement his outsourcing tariff using authorities under current law. (Can he?) Doing so will require coordinating implementation work by Treasury (if it’s a tax), USTR (if it’s a tariff) and Commerce because of the trade and business implications, and State to work on the anticipated blowback from foreign governments whose exports to the U.S. will decline. Part of this would be coordination, part of it internal management to ensure that those within the various departments who recommended against the policy nevertheless work toward implementing the President’s decision.
  • Sales and marketing support: Mr. Cohn may need to visit Capitol Hill to explain this outsourcing tariff and its implementation to Congress, or do press interviews, or meet with business and labor leaders. His staff may also help educate other senior White House advisors and Cabinet secretaries who need to help push this element of the President’s economic policy agenda.

The White House policy council staffs each run a manufacturing shop and a service operation. They help the president manufacture policy decisions and they provide policy services to promote those decisions and coordinate their implementation.

Gary Cohn, Mike Flynn, and an as yet unnamed DPC head will have their hands full.

Thursday, 15 December 2016|

The Trump tariff and the dumb market

Imagine five American firms, each of which lays off New York workers.

Firms 1, 2, and 3 close their New York widget factories.

  • Firm 1 builds a new widget factory in Mexico.
  • Firm 2 builds a new widget factory in South Carolina.
  • Firm 3 does not set up a new factory anywhere. Instead, it buys widgets from a separate company which built a widget factory in Mexico and imported them into the U.S. This separate company never had a U.S. factory.
  • Firm 4 closes its New York call center and lays off all its employees. The firm opens a new call center in the Philippines.
  • Firm 5 keeps its New York widget factory open but replaces half its employees with robots.

In all five cases New York workers lose their jobs. Firms 1 and 4 move New York jobs to foreign countries while Firm 2 moves New York jobs to South Carolina.

President-elect Trump’s proposed new 35 percent tariff would apply to Firm 1, and specifically to goods imported into the U.S. from the new Mexican factory that replaced Firm 1’s now closed New York factory.

It appears his policy would not apply to Firms 2-5. Based largely on his recent interview on Fox News Sunday with Chris Wallace, here is my best read of Mr. Trump’s intent.

  • The policy clearly does not apply to Firm 2, which “moves jobs” within the U.S.
  • He said the policy would apply to a company that “wants to move to Mexico or another country.” Firm 3 isn’t moving anything. Firm 3 shuts down a U.S. factory, while a separate firm makes the replacement goods and imports them.
  • Firm 4 is outsourcing services, not goods. (I think) his policy would apply only to manufactured goods.
  • The tariff also doesn’t apply to Firm 5, since while jobs are lost, no jobs are moving.

Over the next few months Team Trump will have to address the following four questions about his proposed tariff.

Question 1: Will it work? Will the threat of an import tariff prevent the case of Firm 1? In the short run, yes. A punitive tariff can be set high enough that it outweighs the cost advantages of cheaper foreign labor and a less burdensome foreign regulatory environment. The tariff would in effect trap domestic manufacturing capacity and prevent it from moving outside the U.S., and that’s the intent.

At some point, however, another firm without existing U.S. manufacturing workers can set up a factory in Mexico and start making similar goods at lower cost than Firm 1’s trapped U.S. manufacturing capacity. Those goods would not face the import tariff since this separate firm didn’t move jobs out of the U.S. Firm 1 will struggle to compete with these less expensive imports and may eventually shut down its New York factory despite the policy designed to help those workers. In the long run Firm 3 can probably beat Firm 1.

Question 2: Is the Firm 1 case the result of unfair trade? Chris Wallace described Firm 1 as making a free market decision. The president-elect replied “No, that’s the dumb market… I’m a big free trader, but it has to be fair.” Mr. Trump seems to be conflating two things:

  1. a foreign government’s trade policy or a negotiated trade agreement (“bad trade deal”) that disadvantages U.S. producers relative to foreign producers; and
  2. a competitive market advantage to producing goods overseas unrelated to trade policy: things like cheaper labor or resource inputs, or lower tax and regulatory burdens.

The president-elect and his advisors now need to explain why, in the absence of the first, the second is not a free market, why they think it’s unfair trade. If there is something in NAFTA that tilts the playing field away from the U.S. and toward Mexico, I’ve never heard Team Trump explain it. In the case of Mexico, Mr. Trump’s “dumb market” is also a free market, just one in which he doesn’t like the outcome of competition.

Question 3: What share of laid off manufacturing workers see their jobs outsourced to foreign countries? New Yorkers in all five of the above cases are laid off. How many of them are in the first case relative to the others? To answer this you’d need to measure automation vs. outsourcing, domestic vs. foreign outsourcing, and services vs. manufacturing, as well as make a guess about how many firms would choose the Firm 3 path when confronted by a tariff that applies only to Firm 1.

If laid off Firm 1 New Yorkers are only a small portion of all laid off New Yorkers in Firms 1-5, by itself that doesn’t mean you shouldn’t try to help those in Firm 1. It does mean you’ll need to explain why you’re trying to help some laid off workers and not others. That brings us to the political and communications challenge…

Question 4: Is the policy fair and will it be perceived as fair? What do you tell the laid off workers of Firms 2-5 when they ask why you’re not helping them as well? The affected workers, their families, and the local economy, probably care less why New York jobs disappeared than how many did. Team Trump could argue their policy is narrowly tailored to solve only a specific problem, that of manufacturing jobs outsourced overseas because of unfair trade policies or bad trade deals. The laid off New Yorkers probably don’t care whether their jobs were shipped to Mexico, shipped to South Carolina, or taken by robots. If Team Trump can’t answer question 2 convincingly and explain why Firm 1’s workers were harmed by unfair trade policies or agreements, rather than by the harsh realities of the free market on a level playing field, then their justification for helping some workers but not others could fail. And if Firm 1’s laid off New Yorkers are only a small portion of those laid off in all five firms, then Team Trump will face an even greater political and communications challenge.

Rather than an import tariff that may not work in the long run, is unfair to other laid off workers, and undermines free market competition, I’d like to see President-elect Trump dedicate his energy to pushing the other policies he referenced on Sunday: those that make it less costly for firms to employ American workers by lowering tax and regulatory burdens. Make America a great place to invest, expand, and create new jobs. A flexible and rapidly growing U.S. economy is also the best way to help Americans who lose their jobs (for any reason) find new ones quickly.


Tuesday, 13 December 2016|

What are the most powerful policy jobs in the White House?

Yesterday I explained why presidents don’t delegate more policy decision-making power to their cabinet. Instead the president makes the big decisions, supported by his White House staff. This makes the White House staff powerful. Now let’s peek inside the White House.

What are the most powerful policy jobs in the White House?

First, a few caveats.

  • In this post we’ll look only at the top tier of White House staff, each of whom has the rank of Assistant to the President. That’s an oversimplification but a useful starting point.
  • My answer is based on my experience in the George W. Bush (43) White House. Your mileage may vary.
  • The various senior roles have different forms and tools of policy power. I may write about that in a separate post later but won’t do so here.
  • Here I’m focusing on the power that derives from the position and the operating patterns of the White House. Some particularly effective advisors “punch above their weight” and have a policy impact larger than their role might suggest here.
  • I’m using White House a bit loosely. Technically a few of these advisors (OMB, CEA, CEQ, OSTP) are part of the broader Executive Office of the President and aren’t formally in the White House. In practical terms there’s little difference.
  • I’m excluding “policy czars” that existed in the Obama White House but not in Bush 43.

OK, let’s dive in. I’ll divide the senior White House advisor jobs into three buckets:

  1. five policy advisors who run the policy processes within the White House;
  2. six non-policy advisors who have a different principal function but nevertheless play a major role in advising the president on big policy decisions; and
  3. four policy advisors that are the leads for specific areas of expertise.

In terms of total policy impact I rank the buckets 1-2-3. Those in bucket 2, however, often are more powerful than the policy council directors in bucket 1 if we’re thinking about more than just policy impact, and also when we’re talking about tradeoffs among issues.

I’m going to leave the Vice President and the White House Chief of Staff out of the following. Both play at a level above all of what follows.

Bucket 1: The most powerful policy-only jobs in the White House are:

  • Deputy Chief of Staff for Policy
  • Assistant to the President and National Security Advisor
  • Assistant to the President for Economic Policy and Director, National Economic Council (aka “NEC Director”)
  • Assistant to the President for Domestic Policy and Director, Domestic Policy Council (aka “DPC Director”)
  • Director, Office of Management and Budget (aka “OMB Director”)

Within the White House these policy positions have the greatest impact on policy because (a) that’s their job and responsibility, (b) they run the policy decision-making and implementation coordination processes, (c) they work with agencies on policy on a daily basis; and (d) part of their job is to “own” the policy issues within the White House, acting as the keeper of the answer to every question of the form “What is the president’s policy on ________?” In the Bush 43 White House the Deputy Chief of Staff for Policy had an oversight role over the NEC, DPC, and OMB, so you should think of that slot as a notch above those three.

Bucket 2: The non-policy White House jobs that have the biggest policy impact are:

  • Assistant to the President for Legislative Affairs (aka “Leg Affairs”)
  • Senior Advisor or Counselor or Strategic Advisor (the political advisor, e.g., Karl Rove, Valerie Jarrett, and soon Steve Bannon)
  • Assistant to the President for Communication (aka “Communications Director”)
  • Assistant to the President and Press Secretary
  • Chief of Staff to the Vice President
  • Staff Secretary

These six roles each have a non-policy principal function within the White House. They are major players in all the big presidential policy decisions because (a) they are close to the president, (b) they are each the principal advisor on a key element of a president’s policy decision (Congress, politics, communications, press), and (c) they do that for every single policy decision (i.e., maximum breadth). The first five of these six advisors were in every NEC Principals meeting we had, and, more importantly, were in every economic policy decision meeting we had with the president. Bucket 2 advisors often play a smaller role in national security issues because some presidents are reluctant to politicize national security decisions.

The Staff Secretary is a special case with enormous potential to influence policy almost invisibly. I may write about that separately but won’t spend time on it here.

I have excluded the Chief Speechwriter because the role comes with little formal policy power. This, however, is one of those roles where a particularly influential chief speechwriter can occasionally have significant influence on certain policies.

Bucket 3: The White House policy specialists are:

  • White House Counsel (the president’s lawyer)
  • Chairman, Council of Economic Advisers (aka “CEA”, the economist)
  • Chairman, Council on Environmental Quality (aka “CEQ”, the environmental expert)
  • Director, Office of Science and Technology Policy (aka “OSTP”, the scientist)

Each of these is a subject matter expert. Each plays in a wide range of policy issues and has a seat at every policy table where his or her expertise is relevant. These slots are different from the first bucket in two ways. First, these jobs are best held by true experts (e.g., a brilliant legal mind, an accomplished economist and scientist, etc.), where the Bucket 1 policy process management jobs can be held by generalists. The head of CEA has to be a terrific economist, the NEC Director can but doesn’t have to be. Second, these specialist advisors have somewhat narrower subject matter scopes than the policy process managers. For our purpose here we’re looking at the White House Counsel as the legal expert. He or she also has a separate role as the president’s lawyer.

I’ll end by reinforcing yesterday’s post about the policy importance of White House advisors relative to the Cabinet. On national security issues SecState and SecDef are often heavy hitters relative to the White House staff, but on other issues the White House staff’s policy impact often significantly outweighs that of the relevant cabinet secretaries.

Take a straightforward issue like pension reform that involves three cabinet departments: Treasury, Labor, and Commerce. In the Bush White House a policy meeting with the president to get decisions on pension reform would have the following principal-level attendees:

  • Vice President
  • Chief of Staff
  • Deputy Chief of Staff for Policy
  • NEC Director (runs the meeting)
  • Leg Affairs
  • Senior Advisor
  • Communications Director
  • Press Secretary
  • VP’s Chief of Staff
  • CEA Chairman
  • White House Counsel or Deputy Counsel
  • OMB Director
  • Secretary of the Treasury
  • Secretary of Commerce
  • Secretary of Labor

Excluding the president and VP, count ‘em up: three cabinet secretaries and eleven senior White House aides (counting OMB as White House). That partly demonstrates the policy power of White House staff.


Thursday, 8 December 2016|

Why don’t presidents delegate more policy power to their cabinet?

Why don’t presidents delegate more policy decision-making power to their cabinet secretaries? Why does White House staff have so much power over big policy decisions relative to the much more visible Cabinet?

  • Most big policy issues cross multiple jurisdictions within the government, especially outside the national security realm. This makes it hard and at times illogical to entrust one cabinet secretary to make decisions that so directly affect parts of the government for which he is not responsible and does not have expertise.
  • It is quite difficult to get one cabinet Secretary to take orders from another. Cabinet and especially sub-cabinet officials will take and follow orders from White House staff that they will not take from their peers in other departments, because they perceive those White House staff as speaking for the president. Cabinet secretaries are successful people who when hired were told they’d report to the president, not to another cabinet secretary. Not a lot of small-ego type-B personalities here.
  • While you could delegate a certain amount of money to a cabinet secretary, it’s hard to delegate amorphous resources like “political capital” or “legislative priority.” To make these tradeoffs the president looks to those with responsibility for his entire agenda and all of his interests, not just a subset. His White House staff are the only ones with responsibility for that breadth.
  • Cabinet secretaries and their staff have institutional interests that overlap with but differ from the president’s. Treasury staff of course ultimately work for the president and they try to advance his agenda. They also have narrower, more local and self-interested priorities. This leads them to think about the Treasury Department’s issues, problems, responsibilities, powers, and points of view; the same goes for other agencies. White House staff think first, second, and third about what the president needs to succeed.
  • Time and place matter. The senior White House staff sit in the West Wing with the president while the cabinet secretaries are in other buildings, some more than a mile from the White House. The president naturally relies on people close to him. Cabinet secretaries spend little time with the president. White House staff are with / near him every day and have a better sense of what he wants and needs. They are closer when he needs information or to make decisions, and they are better aware of his mindset and perspective.
  • Early in an Administration the senior White House staff slots are filled by campaign aides who are close to the president and are deeply committed to him and his agenda. The president knows, trusts, and relies on them. Cabinet picks are often new hires, and even the best of them take time to earn the trust and respect of their new boss.
  • Some cabinet secretaries are newbies to government, and some just aren’t great decision-makers. Some are chosen for reasons other than their policy expertise or judgment: some are good managers, others are great communicators, and still others are chosen to check political boxes.
  • While there are plenty of cases of White House staff seizing/holding power for their own reasons, in most cases the centralization comes from the president’s desire to make decisions for himself rather than to rely on others, including the Cabinet. It’s the president who centralizes power; his White House staff help him do that. People who run for president tend to like to make the big decisions themselves. They have the ultimate responsibility when things go wrong, so they choose to keep the authority to match that. The buck really does stop at the Resolute desk in the Oval Office.

Update: I struck the opening line about my view of the President-elect’s initial Cabinet picks. It was distracting readers from the main point of the piece.

Wednesday, 7 December 2016|

The costs of protecting 800 American jobs

Last week President-elect Trump announced that he had convinced Indiana-based Carrier not to outsource about 800 U.S. manufacturing jobs to Mexico.
Since then in a series of tweets Mr. Trump has:
  • publicly attacked another firm, Rexnord, for considering outsourcing jobs to Mexico;
  • made clear his approach to Carrier and Rexnord is the beginning of a broader policy;
  • threatened all American firms with “retribution” if they outsource jobs; and
  • proposed a 35% import tariff against U.S. firms that do so.
What, then is so bad about President-elect Trump paying, telling, or forcing Carrier, Rexnord, and American firms generally not to outsource American jobs to Mexico?
About 800 Americans now working for Carrier benefit from the President-elect’s actions. Here are five costs of his approach.
  1. It is unfair to other Americans.
  2. It weakens American firms relative to their foreign competitors.
  3. It does long run economic harm to the U.S.
  4. It abandons any pretense of free trade on a level playing field.
  5. It is easily abused.
President-elect Trump’s approach is unfair to other Americans: consumers, taxpayers, and employees of other firms.
  • Forcing Carrier to pay higher labor costs than they could pay in Mexico will make Carrier gas furnaces more expensive. Some American Carrier employees win, while everyone in the market to buy a gas furnace loses. Many of those losers are American consumers. Mr. Trump is helping a few American workers a lot and hurting many more American consumers a little. These consumer losses increase the more one replicates this policy.
  • Some Americans (Hoosiers in this case) will pay higher taxes to subsidize the wages of others. Why should the employees of Eli Lilly (10,000 Indiana employees), Rolls-Royce and Roche (~4,500 each), and countless other Indiana employers subsidize the wages of Carrier employees because the President singled out this particular firm? That is unfair to those taxpayers.
  • A one-off beating with a government stick is unfair to that firm, while a one-off taxpayer-financed carrot for one firm is unfair to others.
President-elect Trump’s approach weakens American firms relative to their foreign competitors.
  • Mr. Trump’s threatened import tariff applies only to firms that (a) shut down U.S. manufacturing capacity and (b) set up a foreign plant to replace the closed U.S. facility. He would punish movement. Setting aside the difficulty of monitoring this, this disadvantages firms that today employ U.S. workers, since movement is measured relative to your starting point.
  • Any firm that does not now make gas furnaces in the U.S. could set up shop in Mexico, pay lower labor costs than Carrier, avoid an import tariff, and therefore have a significant cost advantage over Carrier. This firm could be U.S-based or foreign. The Trump import tariff would not apply to gas furnaces manufactured by these low cost competitors set up anew in Mexico, only to Carrier if they move capacity there from the U.S.
  • More generally, any manufacturing firm that today employs Americans may now be at a competitive disadvantage, relative both to new firms and firms now employing cheaper foreign labor. If threatened by the Trump Administration, firms employing American workers must either pay higher labor costs than their competitors or face an import tariff their competitors will not face. In the hope of protecting American workers the President-elect is handicapping the firms that employ them.
President-elect Trump’s approach does long run economic harm to the U.S.
  • In attempting to protect the status quo, the President-elect’s threats (tariffs, jawboning, and unspecified other policy sticks) tell business leaders in the U.S. and around the world not to invest in new U.S. manufacturing capacity and not to expand their existing American plants. Why would any firm hire American workers, knowing that they would be penalized if they try to move out of the U.S. at any future time?
  • We can see similar problems in Western Europe. There government policies make it harder for businesses to fire workers. As a result, managers hire fewer workers since they can’t correct their hiring mistakes or lay people off when times are tight. In the short run this looks compassionate but in the long run Europe has much higher unemployment than the U.S. Let’s not replicate the slow growth policy failures of Western Europe.
  • Similarly, if you penalize American firms when they try to lay off employees and shut down manufacturing plants, you will help those employees in the short run but you will get less new investment and create fewer American manufacturing jobs in the long run. That’s bad, and the long-term losses to America are not worth the short-term gains. We should want to expand and attract new investment to the U.S., not just prevent what’s here from leaving.
President-elect Trump’s approach opposes free trade on a level playing field.
  • During the campaign, candidate Trump said he opposed (a) bad trade agreements negotiated by the U.S. government and (b) cheating by foreign governments and foreign firms. Neither is in play here. There is no claim that NAFTA disadvantages Carrier, nor that Carrier’s foreign competitors or Mexico are somehow cheating. Instead Mr. Trump is opposing free and fair competition. The playing field is level, the rules are fair and fairly enforced. Mexican workers are simply less expensive than American workers. By threatening retribution (his word) against firms that outsource American jobs without any mention of bad trade deals or cheating, the President-elect is now embracing straight-up protectionism.
President-elect Trump’s approach is easily abused.
  • Abuse #1 (policy slippery slope) — After successfully pressuring firms not to outsource American workers, why not pressure them not to fire workers for any reason? Why not pressure them to meet other “legitimate” policy goals? This already happens with the force of law through fuel economy requirements, health insurance mandates, and affordable housing goals. The novelty here would be a President pressuring individual firms to meet his own arbitrary policy goals without any democratic process. When you have a hammer everything looks like a nail.
  • Abuse #2 (political selection criteria) — There are too many firms to pressure them all, so how will the new President choose? Over time he or his advisors will be increasingly tempted to pressure firms that employ workers in politically important regions before elections.
  • Abuse #3 (others join the game) — Even if the President does this completely above board, he makes it easier for other elected officials to do the same but with less noble goals. Elected officials already do this with varying degrees of success. When the President does it he signals to other elected officials that this is appropriate and can be effective.
  • Abuse #4 (crony capitalism) — Business leaders will try to curry favor with the President and his advisors so they can nibble on carrots and avoid sticks. Those who like the President and whom he likes will benefit, and vice versa. When a politician rewards his business friends and punishes his business enemies it’s called crony capitalism. It is corrupt and it creates incentives for other business leaders to spend their time and money trying to get similar political access with elected officials. And a firm leader now knows it can initiate a negotiation with the Trump Administration simply by threatening to outsource jobs.
However well-intentioned, President-elect Trump’s new policy will do far more harm than good. I hope he instead embraces free trade and competition and that he stops trying to tell individual American business leaders how to run their companies. He can best strengthen the American economy by developing broad-based neutral policies that create favorable economic conditions and allow smart business leaders to expand, hire, and make the American economy grow.
Monday, 5 December 2016|

I still oppose Donald Trump

In March I wrote:

I will not vote for Donald Trump for the Republican presidential nomination. If he wins the nomination I will not vote for him for President.

This is not a tough call.

Donald Trump is an ignorant, unprincipled, amoral policy lightweight opposed to free market capitalism and limited government.

… Donald Trump is dangerous.

… Donald Trump acts like an eighth grade bully.

… Donald Trump lacks the character, the values, and the sound judgment essential to fulfill this awesome responsibility. He is unqualified and unfit to be President of the United States.

Over the past seven months Mr. Trump has repeatedly reaffirmed my March viewpoint. I continue to oppose Donald Trump and I will not vote for him for President. For me it’s a matter of conscience — I simply cannot vote for him no matter what the alternative or consequence. As I did in March, I again conclude that a President Trump would be a disaster for America, a danger to the world, and destructive of those values and ideas most important to me.

I find that this year I am not thinking of my vote as a choice between two bad alternatives. Maybe I should be; maybe I should vote for whichever major party candidate is less worse for America.

But I cannot. My mindset is more selfish: no one “deserves” my vote, especially if their principal case is a negative one. My vote is mine and you have to earn it by convincing me to vote for you, not just against your opponent. Has Mr. Trump earned my vote? No. OK, then has Secretary Clinton earned it?

Also no. Given my starting point it wouldn’t have taken much. I was gettable. She could have tacked back to the center in the general election rather than continuing to pander to her progressive base. She could have said something positive about free trade, or slowing entitlement spending growth, or keeping taxes low, or reducing the burdens on the private sector of an ever-expanding administrative state. She could have accepted responsibility for her past poor judgment and ethical lapses. She could have taken steps to show that she had learned from these mistakes and that she would not make similar ones in the future. She could have signaled that a Clinton presidency would sometimes be bipartisan, would occasionally be centrist, would always respect the rule of law and behave ethically.

Since she has done none of those things she has not earned my vote.

This means I’m stuck. Donald Trump must not become president but I cannot justify voting for Hillary Clinton. So a write-in it is for me—tomorrow I will write in Jeb Bush as I did in the California primary. I would vote for Clinton if she had tried just a little, so great is the threat that Trump poses. I comfort myself somewhat with this unsatisfying middle ground by remembering that my vote here in California is purely symbolic, and for once I am happy that I don’t live in a swing state.

I am not trying to tell you how to vote, and I respect others who are making different choices. I write this today simply because I want once again to be on record opposing Donald Trump while it still counts.


Monday, 7 November 2016|
Go to Top