The three biggest economic policy problems of our time

As I see it, governments and central banks are currently facing three major economic policy problems.  First, there is the problem of how to manage the economy when interest rates are stuck at zero.  Second, there is the problem of how to deal with rapidly rising debt levels.  Third, there is the problem of how to deal with rising income inequality.  What is interesting is that in each of these problems, public finance economists are likely to play a key role in finding the solution.

When interest rates are stuck at zero, central banks can no longer manage the ebbs and flows of the business cycle by changing interest rates, so they will need a new policy tool to fulfill that role, and one likely candidate is to give central banks some control over fiscal policy.  Central banks can probably muddle through if interest rates remain at zero temporarily, but if this is a permanent problem, then it does probably make sense to start figuring out how the central bank might gain some control over government finances by sending out direct payments to individuals.  This is one key example where public finance economists are going to have to work together with macroeconomists to build a whole new policy framework that gives the central bank the ability to fulfill one of their primary functions over the long term going forward into the future by incorporating tools that come from fiscal policy.

Over the next several decades, governments are also going to have to worry about rising levels of debt whether this is due to dramatic new economic shocks like the housing bubble or the pandemic, or because interest rates are stuck at zero and governments need to run persistent budget deficits just to keep the economy performing at potential.  If the first problem had the public finance economists bailing out the macroeconomists by giving central banks some control over fiscal policy, the second problem has the macroeconomists bailing out the public finance economists by having central banks buy up lots of government debt through quantitative easing.  If interest rates remain stuck at zero as debt levels continue to rise, central banks will likely be doing a whole lot of QE during that time as well, so that even if debt levels go up, much of that debt will be held by central banks.  If central banks continue to roll over that debt in perpetuity and promise to buy more debt if there threatens to be a run on government debt, than it could be the central banks that keep governments from experiencing a debt crisis.  That means on both of these first two problems, public finance economists and macroeconomists are going to have to work together so that both of these highly critical economic problems can successfully get resolved.

In many rich countries (especially the English speaking ones), there has also been a persistent trend of growing income inequality, and public finance economists will likely play a key role in resolving this problem too.  Tax policy will clearly play a big role, as declining marginal tax rates on the rich is likely one of the causes, and increasing marginal tax rates on the rich is one of the likely solutions as well.  Figuring out how to make sure the rich pay their fair share in taxes will likely be left to the tax policy experts, and they might play a key role in deciding how that tax money gets distributed as well, since that could involve expanding tax refunds for low and middle income people.  

That puts public finance economists at the heart of all three of the major economic problems governments and central banks currently face, and I remember the time when I was beginning my career in public policy and had to decide on my application for a research assistant position at Brookings what field I wanted to go into.  I chose budget and tax policy as my preferred field and after I started working on the topic I was always struck how important those issues really were, since tax policy experts routinely dealt with proposals with costs in the trillions and even a billion dollars was just considered a rounding error.  What was especially surprising to me was how few people there were working on tax policy issues at the time, and now public finance economists are even more important than ever.  If people are trying to figure out how to have a big impact on the world, public finance as a field is a good place to start, and tax policy should be one key aspect that young public finance economists should also give special attention to learning. 

Kevin Carey's plan for higher education

Kevin Carey published his plan for reforming our higher education system in the Washington Monthly recently, and there is a lot to like in his proposal.  His plan consists of three major reforms.  First, the federal government would provide a $10,000 per student subsidy for every college that agreed to a transparent pricing system for tuition that made it free for those from families making less than $75,000 and charged only $10,000 for those from families making above $250,000.  Second, he would require any college participating in this system to accept credits from any other college in the system.  Third, he would reduce tax subsidies for charitable donations to colleges with large endowments.  

Biden offers his own plan to provide free tuition at public colleges for students from families making less than $125,000, but his plan has some serious problems that I describe in an earlier blog post, and Keven Carey's plan addresses many of those.  First, Carey's plan would not require states to pony up one-third of the money, which would avoid another patchwork of participating states as happened with the Medicaid expansion in Obamacare.  Second, Carey would have individual colleges decide whether or not to participate, rather than having the state government arrange a deal to have every public college in the state participate.  This would allow some private colleges to participate if they want, and some public colleges to not participate if they prefer that as well, and would get rid of the dynamic where the subsidies for an entire state could hinge on every public college in the state adhering to a number of strict rules.  Third, Carey's plan would get rid of the tuition cliff, where Biden would have students from families making less than $125,000 get free tuition, but students from families making $125,001 could be charged full tuition.  Carey smooths out this transition by ramping up the tuition colleges could charge starting with families making $75,000 all the way up to $250,000.  

One of the advantages of Carey's plan is that it provides a boost in funding to colleges that charge little tuition.  Many colleges charge less than $10,000 in tuition so the $10,000 subsidy would allow them to reduce tuition to zero for everyone if they wanted, and still spend more on a student's education.  At the same time, it would allow expensive colleges to opt out, which would provide for some independence, but comes at the cost of doing little to reduce tuition or making pricing more transparent at the schools that need it the most.  Plus, because those expensive colleges would now be competing with colleges who did get some major subsidies by participating and dramatically reduced tuition as a result, those expensive colleges would face substantial pressures to reduce education spending on their own students in order to attract them to their school.  Clearly, spending more on low tuition colleges is something that is badly needed and often overlooked, but putting price pressure on schools that do not participate might decrease education spending on our best students, which is something we would probably like to avoid.  

I would propose changing his proposal slightly, where schools would either get half of their net tuition revenue in federal subsidies or $10,000 whichever is higher (up to say $25,000), and could charge up to $10,000 or half of their net tuition revenue (up to say $25,000) which ever is higher.  This would encourage some of the more expensive colleges to opt in, which would spread the dramatic tuition discounts to a wider range of schools, and impose some pricing transparency on the schools that need it the most, while also relaxing some of the competitive pressures to reduce spending on those schools who now decide to participate.  The most elite colleges could still opt out and not be subject to the requirements of the federal government, which would still allow for some independence.

As far as Carey's other proposals, making it easier to transfer credits would be very worthwhile, but I am not sure the effect would be quite as transformative as he suggests.  He also proposes eliminating the charitable deduction for donations to colleges with big endowments, and it is useful to point out that Biden's tax plan would reduce the generosity of those deductions already for high income households, which is probably a good idea.  Biden's plan allows for some tax subsidy for donations to colleges, which I also support, but perhaps the full deduction could be allowed for donations to low endowment schools in an effort to spread the wealth beyond a few elite colleges.  Carey would also propose having participating colleges charge the same for in state and out of state students, which is another good idea that is often overlooked.  Charging different prices for in and out of state students distorts decision making among some of our best students (especially for those with few in state options), and provides no overall advantage to the country once every state decides to do it.

In his article, Kevin Carey provides a very useful proposal to reform our higher education system.  It is not perfect, and I would make some tweaks around the edges, but fundamentally, his plan is much more sound than the plan to provide free tuition for those from families making less than $125,000 that Biden proposed.  He specifically addresses some of the major shortfalls of Biden's plan and provides other useful ideas that would make our system of higher education even stronger.  The article is definitely worth a read, and should be getting more attention in our current debate over higher education.

The Triumph of Injustice by Saez and Zucman

I just finished The Triumph of Injustice by Emmanuel Saez and Gabriel Zucman and I have to admit, it is the perfect antidote to this week's breaking story about how Donald Trump managed to successfully avoid paying virtually any federal income tax at all.  The book even leads with a story about how Trump bragged about avoiding taxes in one of the presidential debates with Hillary Clinton in 2016, declaring that this just shows how smart he is.  One of the key points of the book, which I hadn't fully considered myself, points out that one of the way progressive tax systems fall apart is through a surge of avoidance and evasion that causes policymakers to throw up their hands and give up on high tax rates because they appear to be too difficult to collect.  They even argue that tax avoidance gave low tax advocates key leverage when they passed the 1986 Tax Reform Act, because they were able to strike a deal that had all the extra revenue gained from closing loopholes and ending the massive tax avoidance practices go to reducing the tax rates, especially at the top. 

The conclusion that follows is that in order to save our system of progressive taxation and ensure rich people pay their fair share is to reform our tax structure and methods of tax administration in order to end the corrosive tax avoidance and evasion currently plaguing our system.  To their credit, they have a comprehensive plan to make sure this happens that includes reconstructing the way we tax multinational corporations, taxing capital at normal income tax rates, raising the top marginal tax rate, and imposing a new wealth tax on the richest individuals, while also creating a new government agency to regulate tax shelters.  Included in this list are some important new innovations in tax policy that really could improve the way we tax the wealthiest people and corporations.

I'll even admit that they changed my mind on the usefulness of wealth taxes.  I originally thought they were tough to administer, easy to avoid, didn't raise much money, and also encouraged the wealthiest to just move to another country.  Many liberal European countries who have tried to adopt wealth taxes have ultimately gotten rid of them, so that internationally, even in very friendly political environments, the trend has been away from wealth taxes rather than towards them.  They point out that the billionaires in the US have a lot of their income tied up in unrealized capital gains from the companies they founded, and therefore don't report much of any income in a given year, so raising the income tax won't hit them.  Therefore, if you don't have a wealth tax, many billionaires can avoid paying taxes entirely until they die or sell the company, which might take decades and protect them from paying any taxes for most or even all of their lives.

I also have to give them credit for creating an abundance of fascinating tax data, which includes a policy simulation you can use to craft your own ideal tax plan that you can find at taxjusticenow.org.  In addition, one of their first key points shows that the US essentially has a tax system that charges most everyone the same tax rate (between 25% to 30%) except for the ultra wealthy who pay only about 20% in taxes.  They also make the useful point that taxes were much more progressive during and after WW II, even to the point of being deliberately confiscatory, and this still allowed our society to grow and prosper quite successfully. 

I also have to confess that even though I was agreeing with them for virtually the entire book, they did lose me a little bit in the last chapter just before their conclusion.  Here they bad mouth the prospect of creating a VAT in the US, and instead propose an entirely new type of tax, a national income tax, that includes taxes on wages, business profits, and interest income at one flat rate with no deductions.  They argue this is better than a VAT because it is less regressive and has a broader base.  

My reply would be that first we should gradually raise taxes on the top 1%  either until it reaches its revenue maximizing point or until politically it becomes unacceptable to raise them any further.  If we do this first, and then create a VAT, we do not have a progressivity problem because that has already been solved by the other more progressive taxes.   In this case, it does not matter if a VAT even imposes no tax at all on the rich because we have already reached the optimal combined tax rate for that group, and raising taxes even further on that group through a new national income tax would just mean we would need to reduce other progressive taxes to get back to our optimal rates for the rich.  If you are worried about a narrow base, this is largely driven by political pressures and could be a problem in a national income tax as well if political concerns end up overruling ideal policy design.  For me then a better approach is to figure out the taxes on the rich, then raise any additional revenue we need through a VAT and try and solve the narrow base problems by broadening the base on existing payroll, income, and corporate taxes rather than creating a whole new type of income tax.  Plus, given that this type of tax has never been tried before, we should experiment in some smaller states or countries first to work out the kinks, and this could take decades to test and learn how to do right, where a VAT can be implemented right away with few problems off the shelf.

All in all, I came away very impressed with the book, and they provide a tremendous service to those interested in building a just and prosperous society by showing how the rich can and should be taxed.  There is no reason we have to accept increasing levels of tax avoidance and evasion, either from the super rich or multinational corporations, since they have provided a detailed road map as to how we can fix all of these problems and address the issue of income inequality in our country in a very fundamental way.  Progressive taxation it turns out is quite useful and can be saved, and we should move ahead on the task of doing exactly that. 

Claudia Sahm's proposal to make automatic direct payments to individuals in a recession

I just read Claudia Sahm's chapter on direct payments to individuals that was part of a larger book put out by The Hamilton Project at Brookings, called Recession Ready: Fiscal Policies to Stabilize the American Economy.  In it, she proposes a specific automatic trigger based on the unemployment rate to send out cash payments in the midst of a recession.  To her credit, I think she basically nails it.  First, she has Congress and the IRS set up an automatic payments system and the legislation in advance, so the money can go out the door quite quickly.  Second, she creates a reliable and fast acting trigger to get the annual payments out while the recession is still occurring, and third, she also develops a system for multiple payments during severe recessions, that makes sure Congress doesn't take the money away too quickly. 

When I thought about the problem of direct payments, I usually came at the problem from the perspective of the Federal Reserve, where in the last two recessions, interest rates hit zero right after they started, which means the Fed lost their primary tool for managing the ups and downs of the business cycle.  I even suspect that interest rates might stay at zero for very long periods of time, say across the entire business cycle, and thought giving the Fed some power to issue direct payments to individuals would give them a new tool to pursue their counter-cyclical policy mission.  I thought letting the Fed issue direct payments worth $100 per person per month for up to a year, worth about 2% of GDP, at which point Congress could extend them, would be useful.

Examining how the Fed might get some control over fiscal policy was beyond the scope of her paper, so my current view is that I think it makes sense to combine the two proposals.  Congress could pass legislation to set up the automatic unemployment rate trigger, and in one twist, I might give the Fed the power to send out the first annual payment if need be since they might be able to react even faster than a data driven trigger, but if the Fed did not do that, then once the trigger was set off, Congress could vote to up or down whether or not to let the payments go out.  You can debate whether it takes approval from both the House and Senate for the payments to happen, or whether it takes rejection from both the House and Senate (and perhaps the President too) to block them from going out, but I think giving Congress some say in the matter assuages any concerns about giving up too much control over their power of the purse.  Then, once the first annual payment went out, the Fed could decide to provide a monthly direct payment up to $100 per person per month on top of the annual automatic stimulus payment determined by the unemployment rate trigger.  Perhaps after a year of monthly payments, Congress could be given a vote to determine if this should continue or be blocked, just to make sure the Fed doesn't abuse their newfound partial control over fiscal policy.

This would give you the best of both worlds, where individuals would get the benefits of large annual payments on top of a smaller monthly benefit.  Plus, the Fed could make sure the money goes out the door even faster, and adjust the total amount of fiscal stimulus based on the severity of the recession as it happens. I think this balances the need to provide more fast acting and timely counter cyclical fiscal stimulus while still giving Congress some oversight and control over how it takes place.  Given how important the stimulus checks were in this recession, I think it clearly makes sense to have a system already in place to do something similar when the next recession hits.

CBO's long term budget outlook and the impact of interest rates

CBO just came out with their long term budget outlook, and they paint a pretty grim picture.  Federal debt is expected to grow to 195% of GDP by 2050, where the short term spike in deficits due to the pandemic is expected to go away, but the budget imbalance that existed before is expected to continue indefinitely.  Once that existing deficit gets combined with the rising interest costs from the debt (along with rising health care costs), debt rises at an alarming rate for decades into the future.  

My contribution to this debate is to provide some insight into how the path of future interest rates interacts with this long term budget outlook.  The standard way of understanding this connection is to say that if interest rates stay low (say near zero) for a long time, then the costs of servicing the debt goes down over time as well, and this could substantially improve the long term budget outlook.  I would suggest a different problem, however, where if interest rates are stuck at zero, then interest costs might go down, but the primary deficit (the deficit excluding interest costs) needs to stay high in order to keep the economy at full potential.  I estimate then that this need to run high deficits when interest rates are at zero vastly overwhelms any impact on debt service costs, as you can see from the experience of Japan who now has debt as a percent of GDP reaching nearly 240% even though interest rates have been astonishingly low for decades. 

The real wild card in the scenario if interest rates stay stuck at zero is determining what the Federal Reserve will be doing during this time.  If interest rates are stuck at zero, then the Fed will likely be doing a significant amount of QE, which is just buying up a lot of government debt with printed money.  If interest rates stay at zero indefinitely, then the Fed will be doing QE indefinitely, so that even if the federal government is forced to run high deficits every year, the debt that this creates will largely be bought up by the Fed.  This then creates a scenario, much like Japan, where debt levels are very high, but so much of the debt is owned by the central bank that there is more than enough available funds from the private sector to buy up any new debt and hold the existing debt issued by the government, and this combination of printed money from the central bank and excess available private funds makes the debt easy to manage with little fear of a debt crisis.  

That is one way this long term budget outlook could change, where interest rates stay at zero indefinitely for decades into the future.  The other possibility is that interest rates rise above zero again to modest levels, and the Fed stops doing more QE.  In this case, the US will likely be able to reduce the extremely high deficits necessary to deal with the pandemic, but then after that, there will still be a remaining deficit that causes debt as a percent of GDP to continue rising over time.  CBO estimates that in 2025, the deficit will need to be cut by an additional 3.6% of GDP to get debt back to the levels that existed in 2019 before the pandemic by 2050.  This is the real budget danger, where if interest rates rise above zero, then we really do need to cut the deficit, or else debt will continue to grow to alarming levels without the benefit of having the Fed purchase a large proportion of that debt.

The question then is how easily will we be able to achieve that level of deficit reduction.  If a Democrat is president, it is seems perfectly plausible, where both Clinton and Obama enacted changes to the deficit worth more than 6% of GDP.  Clinton took deficits worth 4% of GDP and turned them into surpluses worth 2% of GDP, and Obama took deficits worth 10% of GDP and turned them into deficits worth less than 3% of GDP.  If a Republican is president, then this might prove to be a more difficult task to pull off, where Republicans only seem to be interested in closing the deficit when a Democrat is president, and are perfectly willing to let it grow when a Republican is president.

The real danger then arises not if interest rates stay low in perpetuity, because then the central bank will likely own a good portion of the debt, but if interest rates go above zero, the Fed stops doing QE, and we we elect too many Republican presidents.  These are the kind of things that CBO can't talk about, but are likely going to be the true drivers of long term debt risks, and are also very important to understand.

Improving on Biden's plan to expand Pell Grants

Biden does have a lot to like in his plans for higher education, but I thought it would be useful to focus on a couple of parts that could likely be improved.  Last Friday, I talked about some possible changes to his plan to provide free tuition for all students from families making less than $125,000, and today I thought I would talk about his plan to expand Pell Grants by doubling the maximum amount.

Pell Grants are nice because they are need based and effectively target a lot of money on students who need the most help.  Unfortunately, there are some problems with this program as well.  First, it is overly bureaucratic, where in order to get a Pell Grant students need to fill out a long and complicated FAFSA form, and then they do not find out how much aid they get until after they apply to college and get accepted.  This means that even if it helps students from low income families afford college, the bureaucratic complexity imposes a serious barrier to attending college, especially since they have no idea how much aid they might get when they are making the decision to go to college.  As a result, some studies have shown that Pell Grants have not been effective in encouraging more students to go to college, though they might be more helpful in keeping students in college once they do actually start attending.  

Because of these problems, I would eventually like to reduce our reliance on the federal system of financial aid, and have my own idea about how to provide need based subsidies that accumulate over a child's entire lifetime.  The seed of the idea started with the insight that having the government deposit money into college savings accounts each year as a child grows up (where kids in low income families get significantly more) would be incredibly transparent and easy to understand because students would always know how much is in their accounts well before they decided to go to college as they do in Canada.  This way students would know they have money waiting for them when they graduate, and would also set the expectations that they will eventually go to college early on in their life.  Both features would make it more likely students will decide to attend college, and would be much more administratively simple in some ways since the government could automatically figure out how much to deposit in the account based on income their parents report on their tax returns anyway, thereby requiring no separate form to be filled out.  Plus, benefits would be based on a families lifetime income rather than just one year, so there would be a smaller work disincentive for parents who did qualify for aid.

The problem is that individual accounts are administratively costly, especially when you are allowed to decide how to invest the funds in your accounts, and often have low take up rates, which might especially hurt low income families who have more difficulties signing up.  In addition, it would take decades for a system of college savings accounts to be fully funded, where the benefits for those graduating from high school right now would be quite small.  My key insight is that rather than create college savings accounts for students, instead the government should create a sort of college Social Security plan.  One of the most successful tricks that Social Security pulled off is to have everyone believe all their payroll taxes were saved up and used to pay for their own Social Security benefits, when really all the money they put in was used to fund the benefits for those currently retired.  Similarly, the government could send out statements showing how much a student will have available to them as soon as they graduated from high school, and those statements would start arriving in the very first year a child was born.  The money would not go into individual accounts that could be invested but instead all the money would be used to pay benefits for those currently graduating.  The government could then use existing tax data to calculate how much a student graduating this year would have gotten as if the program had been in place from the very day they were born.  This means those graduating right now would get a full benefit, while at the same time those growing up would know how much they aid they will get when the graduate based on their income so far and would feel like they have a sense of ownership in the system, even though all the money goes to those graduating this year.

The end result would be a kind of system of college pensions, which are actually government funded need based scholarships, where the government puts money into a fund that is used to pay current benefits, but at each point along the way, students growing up are told how much they will get in the future which depends on the income of their parents growing up.  You get the transparency and incentive effects of a college savings account, but the lower administrative costs of SS type system without the decades long delay until the program is fully in effect.  If you gave every child born a $1,000 benefit and then topped that off with $200 a year for kids living with parents in the lowest half of the income distribution (so that a low income student would have about $4,600 to spend on college after they graduated from high school), this would cost about $11 billion a year.  This would allow you to spend some money on expanding Pell Grants, but perhaps the rest of the funds could be used to pay for this special SS style scholarship program, and if this new type of need based scholarships works well, you could gradually divert more and more money from Pell Grants to make the new system more generous over time.  I do think we need to carefully consider how to improve on our current system of financial aid, and this idea offers a new approach and new strategy for providing need based aid that creates a number of important advantages that makes it particularly attractive.

Improving Biden's plan to provide free tuition to everyone making less than $125,000

Today, I'm going to be looking more carefully at one particular part of Biden's higher education plan. In particular, the plan to provide free tuition to all students in families making less than $125,000 a year.  I know this provision is politically popular, especially among college students, but I had a hunch that coming up with an actual plan to do this is actually much more difficult than it initially appears.  To his credit, Biden does have a detailed proposal and actually has links on his website to specific legislation introduced in Congress.  To get a sense of how this works, I actually read the part of the bill with the legislative language that would be used to implement this proposal.  My sense, based on this reading, is that, in its current form, the proposal isn't really workable and would need substantial changes in order to be passed into law.  In particular, I noticed 6 specific problems with the bill.

First, the bill would require states to come up with one-third of all the revenue necessary to fund this plan.  The basic way this plan works is that states go around and get all their state colleges and universities to agree to provide free tuition, and then the states pass legislation to provide one-third of the necessary funding.  States then apply for a grant from the federal government to provide the other two-thirds of the funding and if they meet all the other requirements in the bill, then the federal government will send a check to the states, who will then presumably pass on the money to state colleges and universities, which would reimburse them for all the money lost from eliminating tuition for student from families making less than $125,000 a year.

The basic problem is that it would be very difficult for states to come up with this new funding, especially in these difficult times, so only a patchwork of states would likely participate and in the states that didn't, this generous federal aid could be delayed entirely for many years until they did manage to come up with necessary state portion.  The bill provides about $40 billion a year in federal funds and would require states to provide about $20 billion in federal funds, which would represent approximately a 25% increase in state spending on higher education (unfortunately, I don't have detailed figures for the latest in state spending).  The thing is if states could bump state spending by 25% right now, we wouldn't be talking about the college funding crisis at all, and providing a 2-1 federal match might not be enough for many states to come up with the money themselves (and some states will refuse to participate just to spite Democrats).  Obamacare provided a much more generous match (paying for about 90% of the costs of expanding Medicaid) but many states still have refused to do so.  I'm not really sure why the proposal requires states to provide one-third of the revenue when clearly the federal government is in a better position to provide funding, but by adding this provision it ensures the proposal will get mired down in state budget politics for years to come.

Second, the bill provides insufficient year to year funding growth to replace the cost of lost tuition over the long term.  The way the bill is supposed to work is that in the first year, colleges basically provide free tuition to everyone in families making less than $125,000, and the states and federal government teams up to provide a grant that replaces all the lost revenue.  In future years, this amount is adjusted based on how many people enroll in college, and also by a yearly adjustment.  If colleges end up with more students than they projected, then the funding rises by the increase in the state's GDP deflator (basically an indicator of how much prices went up in the state).  If colleges end up with fewer students than projected, than funding rises by the interest rate of the 5 year Treasury bond.

Clearly, this last part is a mistake.  I think the legislation was designed a few years ago when interest rates were higher.  Right now, the rate on 10 year Treasury bonds is lower than the rate of inflation.  In theory, if student enrollment increases, colleges were supposed to see their funding rise by the amount of enrollment and the amount of inflation in a state, however if student enrollment is declining then the program was supposed to partially offset the loss in funding from declining enrollment by providing a larger increase in state and federal reimbursement for lost tuition revenue by linking it to the rate on 5 year Treasury bonds.  Now that interest rates are lower than inflation, colleges are punished even further when enrollment declines, which is probably not what they intended.

One of the big problems with the entire free tuition approach is that colleges lose a major source of revenue that they can control and replace it with a large government grant that might not grow as fast, so that over time colleges and universities get squeezed.  At first, this might not be a big deal but over the course of decades, colleges might face serious funding problems as their major sources of revenue don't keep up with the rising costs of providing a high quality education.  In theory, if this gets bad enough, colleges could increase teaching loads, which reduces the cross-subsidization of academic research, and research productivity at state colleges and universities could suffer as well.

This legislation basically confirms this fear by only providing an adjustment for enrollment and inflation under normal times when enrollment is increasing.  Clearly, the funding needs to be adjusted for enrollment, but instead of linking the funding increase to inflation, they should link it to overall wage growth.  Colleges spend a ton of money on professor salaries and these costs go up at the very least by the growth in wages and perhaps even more since they need to attract teachers from an extremely talented pool of individuals who are probably seeing even faster wage increases. If this plan is going to be sustainable, they need to make the yearly funding adjustment much more generous when enrollment is increasing (by linking it to wages), and perhaps switch the link to the interest rate on the 5 year Treasury bond when enrollment is decreasing to the growth in wages plus 1% or 2%. 

The third problem with the legislation is that it not only provides for limited year to year inflation adjustments, but also prevents colleges from raising tuition on students not receiving free tuition by more than the year to year adjustment of the overall government grant.  I always thought one advantage to limiting free tuition to those in families making less than $125,000 was that it provided a release valve for colleges, where if the government grants grew too slowly then colleges could just raise tuition on those not covered by the free tuition proposal to make up any gap in funding.  Tuition is already quite a bit lower at state colleges and universities, and high income families have seen a lot of wage growth, so they could probably afford to pay more to send their kids to a public college or university.  By blocking this source of revenue, this just puts an even tighter squeeze on state colleges over the long run when it would actually be quite reasonable for them to raise more money from rich students.

The fourth problem with the legislation is that it allows state colleges and universities to charge tuition on out of state students.  The legislation does limit the cost of tuition on out state students to the marginal cost of instruction, but this in general could end up being quite high, where out of state students could be excluded not only from the benefits of existing state appropriations, but also from the new federal and state funds provided under this program to reduce tuition as well.  I always thought states should charge the same amount to in state and out of state students because otherwise you just punish the ambitious and effectively restrict the choice of colleges for many of them.  This problem is even worse in many small states that might not have a robust system of higher education, where many of their best students really do need the chance to attend public colleges in other states in order to reach their full potential.  Even if it might be unrealistic to expect the legislation to require public colleges to provide free tuition to out of state students too, perhaps they could limit the difference to say $5,000 a year, rather than letting colleges exclude them from the benefits of lower tuition provided by these major sources of state and federal aid. 

The fifth problem with the legislation is that it doesn't provide for any phase-in of tuition for those who are right above the $125,000 cutoff.  The idea seems to be that students from families making less than $125,000 pay no tuition, but students from families making $125,001 would pay full tuition.  Coming from the world of tax policy, large cliffs and cutoffs create bad incentive effects, where if the tuition jumps up too quickly you get some ridiculous marginal tax rates on income, and this could motivate some people to try and game the system by finding ways to reduce their income just below the cutoff.  Realistically, public colleges would have to provide some tuition discount for a dedicated phase-in range, so that maybe if those making less than $125,000 would get free tuition, only those making above say $150,000 or $175,000 would pay full tuition.  The state and federal program only provides funds to replace the lost tuition from those from families making less than $125,000 but doesn't accommodate the need for partial tuition discounts for those from families making just above the cutoff, so some extra funds should be added to make sure this is possible for colleges to try without breaking their budgets in the process.

The sixth problem with the legislation is that it provides a litany of new requirements states and colleges need to meet in order to receive the funding that may or may not really be enforceable.  Not only do colleges need to provide free tuition to students from families making less than $125,000, they also have to maintain the amount they spend on instruction, they cannot deliberately reduce the number of students enrolled, they need to maintain the amount they spend on financial aid, they cannot raise tuition above inflation on students who do not get free tuition, they must have 75% of their teachers be on tenure track, they cannot use grant funds to pay for sport stadiums, merit based scholarship, school administrator salaries, or capital outlays, and states must maintain their previous support for higher education and cannot reduce other spending on need based financial aid.

Now, of course, many of these restrictions have good intentions, but practically things often do not turn out quite as planned.  First, money is fungible.  If you ban grant funds from being used for certain purposes, states and colleges can oftentimes create budget gimmicks to get around them.  Maybe a college wants a new sport stadium, and won't use new grant funds explicitly, but will divert some existing state aid for the stadium from student instruction, and fill in that gap in instruction using new grant funds.  Perhaps states won't be able to explicitly increase tuition, but could jack up room and board fees by quite a bit to raise new revenue.  Its unclear whether this law really will be able to change the behavior of states and colleges, or whether it will just inspire them to create new budget gimmicks.

Second, even if states or colleges do violate these rules, they might be difficult to enforce.  Right now, the federal government only has one very blunt instrument to ensure compliance, and that is denying them access to the federal funds used to reimburse states and colleges for providing free tuition to students from families making less than $125,000.  If one college reduces enrollment by say 2% or only has 74% of instructors on tenure track, the only recourse the federal government has is to withhold all federal funding from their free tuition program for the entire state until the individual school complies.  This is clearly an imperfect way to regulate public colleges, especially since it applies to a long list of new rules and not just one narrow issue like gender discrimination.  Plus, if Biden is President we might be fine, but do we want the next Republican President, who could have a vendetta against the generally liberal institutions of higher education, to have such leverage that they could withhold the equivalent of the vast majority of revenue from tuition for an entire state for some minor infraction?  Virtually every state will have at least one school in violation in some way, and a Republican administration could threaten to block all funding for the state unless they made major concessions in a broad array of unrelated areas to satisfy their demands.  Normally, this scenario would sound implausible, but given today's political reality it might not actually be that unrealistic.  

Third, it is unclear whether we want these restriction to be carefully adhered to over the long term.  Essentially, by blocking changes in tuition, enrollment, need based funding, merit based funding, state policies on higher education are locked in just where they were before the free tuition policy took effect.  Over the short term this might not be a big deal, but states over the long run would probably like to be able to adjust their policies eventually, and not be micromanaged at every step by the federal government under this new subsidy plan.  In general, the states have been more successful designing useful higher education subsidies and implementing them effectively than the federal government, where both Pell Grants and federal tax credits have turned out to be overly bureaucratic, poorly understood by the public, and done little to actually increase enrollment in college.  Giving the federal government the opportunity to micromanage every aspect of state higher education policy might turn out to be ill advised if states can make better decisions on their own.

Practically then, I have offered some important substantive suggestions on how to improve Biden's plan to provide free tuition to every student from families making less than $125,000.  Ideally, however, I would adopt an entirely different approach, where if the federal government wants to spend $40 billion on higher education subsidies, and states currently spend approximately $80 billion a year themselves (again I don't have precise current figures), then the federal government should just offer a 50% match for every dollar the state spends on higher ed, and let them decide how to use it.  States might not lower tuition down to zero, but they would likely reduce it substantially, and giving them some flexibility on how to use the funds might encourage new innovations and approaches that turn out to be even more successful, and this flexibility would be especially important amidst the financial chaos of the pandemic.  This way, the federal government wouldn't be micromanaging state higher education policy and colleges wouldn't get squeezed long term with insufficient funding, so you do get substantial upsides of better affordability and lower tuition without the major downsides that could completely upend higher education in the US.  I think this approach merits serious consideration, especially given that Biden's current plan has some significant problems, and I hope this blog post highlights some ways that higher education policy could be made better if Biden wins the election in November.

What is the likely impact of Biden's budget proposals?

The Penn-Wharton Budget Model recently came out with a useful and highly informative analysis of Biden's budget proposals.  The short version is that Biden does spend quite a bit of new money on education, public investment, housing, and Social Security, but this is basically offset over the first 10 years from the taxes they raise and the savings from their health care proposals.  After 10 years, the extra public investment in infrastructure expires, so that over future decades the taxes and health care savings end up reducing the overall level of federal debt over that time.  

There is a little quirk, where the headline analysis says that Biden spends $5.4 trillion and raises $3.4 trillion in taxes, indicating a gap of $2 trillion over 10 years, but that this only increases federal debt by a tiny amount of 0.1% by 2030, which only represents an increase of about $20 billion.  I think the way this gets resolved is that Biden plans to lower the Medicare eligibility age from 65 to 60, which increases Medicare spending by the government, but decreases spending by private insurance, lowering health insurance premiums for everyone.  This in turn will reduce health care spending by businesses, which raises the taxable income of workers, since they get to keep a larger portion of their income and money spent on health insurance is not taxable but the take home wages of workers is.  This leads to a lot more tax revenue over time, but doesn't show up in the headline revenue for his tax plan since it isn't one of his formal tax increases.

That means the big takeaway is that even if the debt as a percent of GDP was growing before the pandemic, and went up dramatically because of the pandemic, Biden's proposals over the first 10 years do not make this any worse, and if his infrastructure investments end after only 10 years, then the long run budget situation is made a lot better over the following decades.  The US still has a lot of fiscal space to increase their debt over the short run, and the we should be willing to spend whatever is necessary to get us through the pandemic, but I think it is worth noting that we might need to run significant budget deficits in perpetuity if interest rates stay stuck at zero (as has happened in Japan) so we need to think about what we might do if this development does actually occur.  The answer is not austerity, but perhaps instead making our temporary QE permanent, which might happen automatically if we need to keep doing QE in perpetuity as well.

Again, Biden's plans do no harm in the first 10 years and probably help things over the subsequent decades, so I think it is a very reasonable and responsible budget proposal for our current times.  Plus, liberal economists have been arguing for a long time that we should be increasing public investment now, even if we need to borrow in order to do so, since the benefits pay off over the long run and it is very cheap to borrow more money.  Biden's plan finds a way to do that investment without raising debt significantly over the first 10 years and causes it to decline later on.  Add to that the valuable spending proposals paid for by the top 1%, and there is a lot to like in Biden's budget plan.  

enterprise

enterprise - n. 1. An undertaking, especially one of some scope, complication, and risk.  2. A business organization. 3. Industrious, systematic activity, especially when directed toward profit. 4. Willingness to undertake new ventures; initiative.

American Heritage Dictionary, 4th Edition

What happens after we get the vaccine?

So, I'm an economist and not a public health expert, but I still like to try and game out how things might go in the future just for fun.  For the record, I've pretty consistently underestimated the severity of the virus, and it just always seems to turn out a lot worse than I ever thought realistically possible.  That said, I'm not currently as worried about what happens once we get the vaccine as I was before, and here is why.

The first thing I was slow to realize is that a certain percentage of the public will have immunity because they already had the virus and have now recovered.  In NY, they estimated this to be around 20% of the population based on serology tests, which is about what you get for the entire US right now if you multiply the number of confirmed cases by 10.  By the time we get a vaccine, this number could be around 30% with immunity if we continue to get about 40,000 new cases a day.  

Infectious disease experts seem to say that we need to get 60% to 70% of the public to be protected in order to get herd immunity, so if 20-30% have immunity from having had the disease already, that means we need about half of those remaining to be protected from the virus.  Once we have a vaccine, this level seems possible to achieve even without everyone getting vaccinated.  If the vaccine is say 70% effective, then only about 70% of the public would need to get vaccinated in order to reach herd immunity.  If we reach this point, then great, we have herd immunity and the virus should quickly fade away entirely.  However, if the vaccine is less than 70% effective or less than 70% of the public get vaccinated, then we might not have herd immunity.  

If we end up in a situation where there is not herd immunity, I see three possible outcomes.  First, we could see the number of new cases decline substantially as vaccination rates go up, only to have social distancing efforts get relaxed once everyone thinks we're in the clear.  Once we attempt to go back to normal, then we realize we haven't actually reached herd immunity and the cases start increasing again.  This will cause us to cycle between the two modes of high restrictions and declining cases and loose restrictions and increasing cases.  

The second possible outcome is that once we realize we don't have herd immunity we make a big push to increase vaccination rates, perhaps through social pressure, financial incentives, or as a last resort mandatory vaccination rules.  In this case, we might not get herd immunity right away, but do get there a little bit later through increased vaccination rates.  

The third possible outcome is that as we cycle back and forth between tight restrictions and loose restrictions more and more people will get and recover from the disease which then gives them some immunity after they recover.  Even if we assume a worse case scenario, where say we start with 25% immunity, the vaccine is only 50% effective, and only 50% get the vaccine, then we will still have 50% of the public protected and it might only take an additional 10% of the public to get the disease to finally achieve herd immunity.  This is less than ideal, but not necessarily horrific either.

My best guess is that the number of new cases will start to go down as more and more people get vaccinated, and this will spur us to open up too quickly.  Once cases start rising again, we will realize that we don't have herd immunity and more people will feel inclined to get vaccinated, especially since there will have already been an initial round of vaccinations, so we will have a good sense of whether or not the vaccine is safe. Even if we don't get that, then we just cycle back and forth for a bit until enough people recover from the disease to push us up to herd immunity, but that won't require too many to get infected since a large portion will be protected already, either from a vaccine or having had the virus already before the vaccine was distributed.  

Of course, the key is to get as many people vaccinated as possible, and there will be a great political divide over this question, but my takeaway is that even if we don't get everyone vaccinated, the consequences won't be too dire.  It is still perfectly possible to achieve herd immunity without perfect vaccination rates, and even if we only get close, it won't take that much to push us over the top.  That's a lot more encouraging than apocalyptic visions of a new intense political conflict over vaccines that makes the debate over masks look like child's play.  Herd immunity might not happen right away, but we will eventually get there, which gives us some hope that this pandemic really is a temporary phenomenon.  

The importance of financial self interest in politics

A while back, I tried to construct a simple model of politics where everyone just voted for whoever promised them the most money.  If you assume the government imposes a linear tax rate and offers a lump sum benefit to everyone, that means the population with lower than average income will always want to vote for politicians proposing a larger government (since they get more in benefits than they pay in taxes) and half the population with higher than average income will want to vote for politicians proposing a smaller government (since they pay more in taxes than they get in benefits).  What is interesting that in this simple set up, the swing voter with an average income will be completely indifferent to proposals to make the overall size of government larger or smaller (since they pay the same amount in taxes as they get in benefits), so they vote for other reasons entirely, ideally perhaps competence, experience, or what is best for the country, though this could also be for higher deficits, identity politics, or subsidies targeted directly at them.

This is why politics is so stable in developed countries.  Parties will divide along this size of government question and around half will support one party and around half will support the other, and if the other factors like competence, experience, or what is best for the country varies from election to election across both parties, than each party will win and get the chance to govern about half the time.  If politics organizes around identity politics, these voting shares tend to remain fixed over time, and one party who has the majority identity will consistently win and exclude the other party from governing all together. This is what happens in developing countries a lot, where there is not a large robust middle class.

The real threat to our politics then is that it will realign entirely around identity politics and financial self interest will no longer be the dominating factor in voting.  This could lead to one party remaining in charge for long periods of time and a lack of political competition overall, that causes governing to deteriorate over time.  Its not clear which party benefits from the identity politics, where whites are still a majority in the country, but the share of minorities is growing over time, and many white people reject a politics that tries to divide the country by race.  Keeping politics oriented around personal financial self interest, however, is what has kept politics stable throughout the western world for many decades and upsetting this balance is a dangerous development in politics. Appealing to personal financial gain, even if it seems a bit selfish, could actually lie at the heart of what keeps our democracy stable for decades into the future.  

Why do interest rates keep falling?

One of the most important macroeconomic developments over the past few decades has been the continual decline in interest rates.  In the early 1980s, the federal funds rate peaked around 20% and then after the financial crisis interest rates remained stuck near zero for about 7 years.  A variety of explanations have been given as the cause of this development, but my own personal theory is that we have simply started running out of useful investment opportunities.

The basic argument goes something like this.  Back at the beginning of the industrial revolution the amount of physical capital existing in society was quite low.  Over the subsequent century that means there was a wealth of useful investment opportunities that saw the amount of capital gradually increase decade after decade.  My theory is that there is basically a limit to the amount of useful capital accumulation.  If you look at human capital, there is a limit to how much education is really useful, where even if we might eventually get everyone to 25 years of schooling, giving everyone say 50 or 60 years of schooling just doesn't make sense.  Similarly, back in 1900 nobody owned a car and over the course of the 20th century we made it the point where most everyone has 2 cars, but at some point we aren't going to need 5 or 6 cars per family.  

The practical way this develops in an economy is that at first, with a limited amount of capital to invest, the most valuable investment projects are done first.  If interest rates start at say 7%, then you invest in all the projects with a rate of return above 7%, and then once those are done interest rates fall to 6% and you do all the projects with a rate of return above 6%.  This process keeps happening until interest rates keep falling and eventually you reach the point where we have enough capital to offset all the depreciation, plus invest in all the new technologies that are developed, and don't need to do any additional capital accumulation. When this happens, interest rates fall to zero and you still have excess savings to invest with nowhere to go.

If this were the reason for the low interest rates, then this is likely going to be a permanent problem.  At this point, interest rates in the US have fallen to the point that they hit zero in the trough of the business cycle, but still go above zero in the peak.  In the latest business cycle in Japan and Europe, interest rates have remained near zero throughout the entire cycle, and its perfectly possible this might be true for the US in the current business cycle and every business cycle after that.  In that case, we might be facing the real possibility that interest rates are going to remain at zero in perpetuity, and should restructure our institutions (perhaps by giving the Fed some control over fiscal policy) so that we can deal with such a development.  Most economists still expect to have interest rates rise above zero in the US, but just in case they're wrong, we do need to start planning for world of interest rates that stay near zero forever.

My first twitter war

I got into my first twitter war with Dean Baker today, and it was all over the state and local tax deduction.  As Christopher Pulliam and Richard Reeves from Brookings pointed out in their recent report, the deduction is just a boon to the rich (which it is), but Dean Baker responded that raising taxes on the rich in high tax states might cause them to move to lower tax states.

To be honest, for a long time, I thought the deduction was potentially beneficial since it offsets some of the taxes paid at the state level, making it easier for them to raise taxes and provide more funding for valuable programs.  Practically, however, whenever state level politicians start arguing over higher taxes, they basically never bring up the fact that raising taxes on the rich will be partially offset through lower federal taxes because of the state and local tax deduction. This could be an important subsidy for higher state income taxes, but its mechanism is so complex that no one really picks up on it enough to have an impact on state policy.  Ultimately, then I had to reject this argument for the state and local tax deduction.

Dean Baker offers a different argument for the state and local tax deduction in that it discourages high income taxpayers from moving to low tax states.  You hear this argument all the time from conservatives whenever someone proposes higher taxes at the state level.  I remember back when Jesse Ventura was governor of MN, and he made a visit to some MN transplants in Florida, and they all said they moved there because of our horribly high tax rates, so Jesse felt the need to propose a big tax cut in the state income tax.  Of course, Minnesota's extremely difficult winters had nothing to do with their decision to move to Florida, and they only decided to move their *after* they had retired, but this argument keeps coming up all the time when discussing state taxes, and I was a little surprised to hear it from Dean Baker.  

As it turns out, Michael Mazerov at CBPP did an excellent job responding to this argument back in 2014, so I posted a link to his report in a reply tweet.  Dean Baker said that this report didn't refute his argument since it didn't address *changes* in the top income tax rate.  I pointed out that the CBPP report (in Appendix 2) cited two studies that found a negligible impact on interstate migration when NJ and CA raised income tax rates on the wealthy in their state.  He replied that these were only short term studies and you wouldn't expect people to move in only a year.  I responded that one of the studies looked at migration patters 4 years before and 4 years after, so you would expect to see any impact show up in the data by that time.  That's where things currently stand. 

Now I understand that the Democrats face a political problem.  The $10,000 limit on the state and local tax deduction does predominantly impact people living in blue states, and politicians do need to be responsive to their constituents.  This is actually less of a problem after the big Trump tax cut, because that bill almost doubled the standard deduction, which means that now only about 10% of the population itemize their deductions now.  True, these families in the top 10% are now key swing suburban voters, and Democrats need every vote they can get right now, but eventually, down the road, when our democracy is not at stake, I think it does make sense to get rid of the state and local tax deduction.  It does primarily impact the rich, it doesn't really encourage more state spending, and the money raised could be used to directly pay for valuable programs either at the state or federal level.  Pointing out these arguments is a valuable service that Christopher Pulliam and Richard Reeves at Brookings provided for us, and I thought that was worth highlighting.  

How should we phase down the extra unemployment benefits?

On Friday, Jason Furman threw down the gauntlet on Twitter, challenging those proposing automatic stabilizers to come up with a plan to phase down the extra unemployment insurance benefits.  He argued that $400 a month was not unreasonable given the latest unemployment report putting the unemployment rate at about 8.4%, and suggested that if you are arguing for $600 a month now, then that means $600 a month was far too low when the unemployment rate was much higher a few months ago.  Opponents argued that the latest research shows that the extra unemployment benefits had no impact on employment, so we should keep the full $600 in benefits even though the unemployment rate is much lower. 

After considering these concerns for a bit, I think I squared the circle in this debate.  I think it is true the research shows that the extra unemployment benefits have no impact on employment, and this holds true even when unemployment is at its current levels.   We know this because the dramatic cut in extra benefits down to zero for August did not lead to a surge in employment during the month, where gains in employment actually slowed compared to the month before.  This would seem to imply that right now, in the midst of a pandemic, there really is no trade off between more generous unemployment insurance benefits and employment, so we should just set the benefits at the level we need to get the optimal amount of social insurance right now. 

Ideally, this means we would set the unemployment insurance benefits so that they would replace 95% to 100% of income, but the state unemployment insurance systems aren't sophisticated enough to do that, so we're kind of stuck setting it at a fixed amount for everyone.  If we do that, then we'll have to give a bit too much to low income people and not quite enough to high income people, but Democrats believe in redistributing income to the poor anyway, so they would probably prefer to err on the high side.  Some research has indicated that an extra $600 a month in benefits means that about two-thirds of the unemployed will get more in benefits than they earn from their job, and this doesn't seem like an unreasonable number given that low income workers have been hit the hardest as a result of the virus.  Plus, there is some evidence that the percent of households feeling food insecure has risen since the pandemic began, so this would indicate that perhaps we aren't doing enough.  

My contribution to this debate is to argue that *while the pandemic continues* there is no trade off between higher benefits and employment so we should offer the amount the provides the optimal level of social insurance, which is probably around $600 a month.  Once the pandemic goes away, however, then we are back to the normal situation where we want to encourage more people to go to work as the unemployment rate declines, since more jobs are becoming available.  To accommodate both considerations, I would set some threshold level for the virus, where perhaps as long as the number of new daily cases in a state stays above 1 for every 100,000 in population (or about 3,000 new cases for the whole country), then we continue to provide the full $600 a month in extra benefits.  Then, once the virus is under control, we can link the extra unemployment benefits to a specific unemployment rate.  For example, an unemployment rate above 11% in a state would garner $600 a week in extra benefits, and rate between 10% and 11% would lead to benefits around $500 a week. This would keep going down so that benefits would fall to $400 for 9% to 10%, $300 for 8% to 9%, $200 for 7% to 8%, and $100 for 6% to 7%.  You could tweak the numbers a bit so maybe the benefits phased down from 5% to 10% rather than 6% to 11%, but that's generally how I would think about phasing the extra benefits down. 

I think the main point here is that we want to make sure we get the incentives right.  For now, the highest priority should be getting the virus under control, and only once that problem is solved do we want to create incentives to get back to work.  Until then, we need to make sure people can meet all their needs through unemployment insurance benefits alone, and given that we are stuck with a fixed amount of extra benefits for everyone, $600 a week seems reasonable.  True, some people will get more than they earned working, but there are signs that there is still an unmet need even with extra benefits at $600 a week, and the main argument against it, that it discourages people from returning to work, is refuted by the latest evidence available.  I think this proposal balances the arguments coming from both sides, and strikes a nice middle ground that can guide us on how to proceed going forward into the future.  

Should Biden limit taxes increase to those making more than $400k?

Biden recently announced that no one earning less than $400,000 would see their taxes go up, and this approximately corresponds with people in the top 1%.  Given the incredible stakes of the election, this cutoff makes sense politically, especially since one of the swing constituencies this year is the relatively high income people living in the suburbs.  Reassuring those people that their taxes won't go up will help win those votes.

Targeting tax increase to the top 1% also makes sense for trying to solve the income inequality problem.  The share of income going to the top 1% rose from 10% of GDP to 20% of GDP, so clawing some of that back through higher taxes makes a lot of sense.  Given that Biden is a moderate and has some plans for new spending, but not exorbitantly costly plans, limiting tax increases on the top 1% should be more than enough, and could easily raise taxes worth 1% to 2% of GDP.  

Long term, however, this income threshold for tax increases is far too high.  At some point, I would like to get rid of some more of the itemized deductions still left in the tax code, and raise the income cap on the payroll tax for Social Security.  This would generally impact only the top 10%, which still protects a lot of potential voters, and targets the tax increases to those doing relatively well in this economy.  

Eventually, if you want to do a lot more spending, you do start to run out of money by limiting taxes on the top 1%.  Once we get the tax levels on the top 1% to the place we want them, I would propose a grand bargain, where Republicans let Congress enact a new Value Added Tax (at say a rate of 6%) and then the Democrats allow Congress to repeal the payroll taxes used to fund health care, plus the extra tax on investment income enacted along with Obamacare, while keeping the payroll taxes funding Social Security untouched.  This would give a tax break to the rich in a variety of ways, and would allow Democrats to raise an extra $100 billion a year or so to spend however they want.  This would move us closer to the grand bargain often used in Europe, where governments do enact generous expansions of the social safety net, but usually use broad based tax increases like a Value Added Tax to pay for it.

So Biden's plans are just fine for now, and do make sense in this difficult political environment, but down the road, I think we do need to look at other options for raising taxes, and I do worry a little bit about setting the precedent that only the top 1% should pay higher taxes long term.  

New CBO report on the budget outlook

The CBO just came out with a new update to their budget outlook.  The bad news is that the budget deficit is expected to rise to $3.3 trillion in 2020 or about 16% of GDP, which is high.  The good news is that the deficit basically returns to pre-crisis levels by 2023, and that interest costs long term are lower despite higher debt because of lower interest rates.  Debt held by the public is expected to grow to 98% of GDP in 2020 and up to 107% of GDP by 2023.  

One important thing to note (and I had to look this up) is that debt held by the public does not include debt held by the Federal Reserve, and they have been buying up enormous amounts of Treasury bonds.  They announced a $500 billion dollar purchase of Treasury bonds on March 15th as well as an open ended purchase of Treasury bonds that was not widely publicized later on that month, and then they said they would buy at least $80 billion a month (or about $1 trillion a year) after that.  That means a lot of that $3.3 trillion dollar deficit is going to be covered by bond purchases from the Fed, though its unclear exactly how much that might be, and its perfectly possible that the Treasury debt purchased by other investors and institutions besides the Fed might be just about as much in 2020 as it was in 2019.  The deficit is projected to be about $2.3 trillion higher in 2020 than in 2019, but the Fed increased their purchases of Treasury debt by at least $1 trillion this year and perhaps even as much as $2 trillion. 

The takeaway is that even without a lot of Fed purchases the budget will be OK in a few years and in even better shape long term because of lower interest costs.  Plus, if the Fed keeps buying tons of Treasury debt this makes the deficit problem even less concerning, so that debt worries should not be limiting what we spend on getting our way through the economic difficulties created by the pandemic.

Do taxes need to rise to pay for our new debt?

Simon Wren-Lewis has a nice summary of a debate between Jonathon Portes and Bill Mitchell about whether we will need to raise taxes in order pay for the new debt accumulated during the pandemic or to pay for an expansion of public spending in order to pay for a variety of possible priorities.  The quick summary is that both agree we do not need to raise taxes to pay for the debt accumulated during the pandemic, that this debt should allowed to shrink as a share of GDP as economic growth continues over the next couple of decades, and does not need to be formally repaid by running surpluses for a long period of time.

Even if they agree on this point, they disagree whether an increase in public spending needs to be paid for with new taxes.  Jonathon Portes seems to think that over the medium term if we do not raise taxes to pay for it inflation will increase and it is better to raise taxes to avoid the higher inflation then to suffer through persistently higher interest rates.  Bill Mitchell seems to think new public investment will not raise inflation over the medium term so there is no need to raise taxes.  He does eventually admit that taxes might have to go up if public spending rises dramatically and inflation does rise, so there might not actually be much of a disagreement here.

Simon Wren-Lewis goes on, however, to make two key points.  First, he says we should wait to raise taxes until interest rates start going up.  If interest rates stay at zero, then there is no trade off between higher taxes and inflation or higher taxes and lower interest rates, and more fiscal stimulus is necessary to keep the economy at full employment.  It is only when interest rates rise above zero that we know it is safe to start withdrawing the fiscal stimulus, so this first point by Simon Wren Lewis is well taken.

Second, he argues that when interest rates are stuck at zero then fiscal stimulus should be the primary tool of macroeconomic stabilization.  This too seems kind of obvious.  Central banks usually use interest rates to do macroeconomic stabilization but when interest rates are stuck at zero that tool is no longer available.  Fiscal stimulus then remains the main tool available to fill in on that role, but what Simon Wren-Lewis does not do is suggest that perhaps central banks should be given some direct control over fiscal policy since institutionally legislative bodies are not well set up for this particular purpose.  Of course this has to be done incrementally with careful safeguards put in about the extent and duration of the stimulus, but the longer interest rates stay stuck at zero, the more seriously we need to consider this option.  

probity

probity - n. Complete and confirmed integrity; uprightness: "He was a gentlemanly Georgian, a person of early American probity" (Mary McGrory).

American Heritage Dictionary, 4th Edition.

All hail the glories of Kant

Kant has long been my favorite philosopher, so I was excited for the chance to read the long and detailed biography by Manfred Kuehn. From a biographical perspective, Kant lived the quiet life of an academic.  He was single his entire life, and did not produce any major works until his mid-50s. He started out as an adjunct professor and only became a full professor in his mid-40s.  Throughout his career he spent much of his time lecturing for his students and working on his books and articles, while also enjoying a healthy social life with his close friends. 

Intellectually, of course, he was a powerhouse.  He produced three major critiques, The Critique of Pure Reason, the Critique of Practical Reason, and the Critique of Judgement.  He produced a major work of moral philosophy, The Groundwork for the Metaphysics of Morals.  He produced an important work on the philosophy of religion, Religion Within the Limits of Reason Alone.  Finally, he also produced a highly influential essay on how best to achieve world peace called Perpetual Peace: A Philosophical Sketch.

His first major work was the Critique of Pure Reason.  In this work, he attempts to answer the critical question of epistemology, "What can I know?" or perhaps more precisely, "What can I know a priori and in complete isolation from experience?"  In this pivotal book, he rejects the idea that we can make substantive claims about the world independent of experience.  In one of his key insights, he does see knowledge as having an a priori component, but this a priori component is based on the structure of our cognitive apparatus that provides an overall framework that allows us to understand our sensual experiences. These insights then into how the world works are not independent of our experiences but instead depend practically on the structure of our cognitive apparatus.  This leads Kant to reject the traditional a priori proofs that were big at the time about the nature of the soul, the world as a whole, and about God, and causes him to conclude that even though we must assume that God exists in order to maintain a functioning moral framework, we cannot prove he exists.     

This last theme is also taken up in his second major critique, the Critique of Practical Reason.  In this book, he suggests that people can only achieve true freedom through the experience of morality, and that it is only when everyone lives morally that freedom begins to be enjoyed by society.  Furthermore, happiness and morality do not always go together, so then the only way for society to achieve true freedom is to create this link between happiness and morality over the long run by assuming that God and immortality exists.  It is only through these assumptions that the whole moral undertaking of society can succeed, and so therefore we must believe in their reality on this account.

In Kant's third critique, The Critique of Judgement, he spends the first part discussing the problem of aesthetics and details some ideas on the nature of beauty and the sublime, and in the second part discusses the problem of determining whether features of nature can be explained entirely through pure mechanical law.  He argues that "All production in nature must be judged as being possible on mere mechanical law", and that this does not contradict the idea that "Some production of such things cannot be judged as possible on mere mechanical laws."  In his view, each statement has a role in science, even though the second one should be used sparingly.  

In Kant's primary work on moral philosophy, The Groundwork of the Metaphysics of Morals, he argues that moral behavior is not determined by consequences but instead by intentions, and a truly good will only acts in accordance with the categorical imperative which states that "I ought never to act in except in such a way that I could also will that my maxim should become a universal law."  An act in accordance with this categorical imperative is one that is moral but only if done for the sake of duty and not for the sake of self interest.  Many of our acts do appear moral but are in fact done out of self interest, and the only truly moral acts are the ones done for the sake of duty, which makes the intentions of the act the key feature of moral behavior.

Kant also produced a major book on religion called Religion Within the Limits of Reason Alone. In this book, he begins by acknowledging that humans are prone to evil, where even if we understand what the right thing to do is, we still often do what is wrong.  This propensity however is not definitive and humans do have the capability for moral improvement.  The question then is how to achieve this improvement.  Kant believes it is irrelevant whether the temptation to do evil is within us or without us, and that the stories about Jesus help us understand our moral predicament but do not represent anything real outside of the moral lessons.  Kant then goes on to address the question of the possibility for social progress, and the creation of an ideal society where human beings are motivated not by coercion from the state but by their own sense of virtue.  In this sense, he believes that this can be achieved by a community motivated by a faith in God, but only in a purely moral sense and not based on the historical, revealed, or culturally based features of the church.  In his view, it is this rationally based moral component of religious belief that is most important and that any service to God over and above "good life conduct" is "mere religious delusion and counterfeit service to God."  This direct attack on the traditional religious functions of the church got Kant in trouble with the religiously oriented government to the point that he was eventually forced to refrain from commenting on religious matters entirely.  

In a smaller essay, Kant described the possibility of achieving perpetual peace in world affairs in Perpetual Peace: A Philosophical Sketch.  Most famously, he argues that perpetual peace requires a republican or representative form of government that is consistent in many ways with what we know as democratic rule today.  He also proposes that states follow certain rules that are supposed to regulate international relations like there should be no peace treaty with a secret reservation of material for another war and that states are not the kinds of things that can be acquired by other states.  This article is still taught today as one of the early examples of arguments in favor of the notion that world peace can be achieved by making changes to the domestic political structures of countries around the world. 

Kant then was extremely intellectually ambitious and made several tremendous contributions to philosophy that still resonate hundreds of years later.  He was an enlightenment philosopher that believed reason could improve the lives of people over time.  He greatly enhanced our understanding of epistemology by refuting the idea that reason separate from experience could inform us about how the world actually worked.  He emphasized the importance of intentions in morality and how we really should act as if everything we did was actually a universal law.  He emphasized the role of reason in religion and how it was morality that made religion important and not religion that made morality important.  Finally, he set out a path about how world peace could be achieved over time by making reforms to our domestic political institutions.  

By making these intellectual breakthroughs he laid out a path for social progress to follow, and many of these goals, dreams, and ideals he described over 200 years ago are now on their way to being achieved.  For this he deserves to be celebrated, and thankfully he is still taught in colleges across the world today.

What is the Fed doing differently on inflation now?

The Fed made a variety of changes to their monetary policy framework, but the one that has gotten the most attention has been their shift on how they approach inflation.  Instead of strictly targeting an inflation rate of 2%, they are now targeting an average inflation rate of 2% over the long term. The easiest way to think about this is that the Fed is trying to reduce the long term fluctuations in the inflation rate, where now if inflation falls below 2%, instead of targeting exactly 2% the following year, they will target something slightly higher than 2% to help make up the difference.  

In the short run, this leads to larger fluctuations, where if inflation is too low one year, then it will be higher the next, but because these short term fluctuations are designed to offset one another, the long run fluctuations in inflation will be more subdued.  The trick is that you do not want to make up the difference too quickly in order to avoid wild short term swings in inflation, which is why the Fed is adopting a "flexible form of average inflation targeting", which means they'll use their own discretion as to how to make up for the previous deviations from target, which I think makes sense.

This is in fact what the engineering department tells us we should be doing, where the debate over targeting the inflation rate versus the price level is basically a false choice.  Instead of doing inflation targeting *or* price level targeting, you should primarily be targeting the inflation rate with a little bit of effort put into trying to control the price level, which is what control theory engineers do when they use PI (proportional-integral) control, which I explain in more detail in one of my first blog posts.  The new Fed target does exactly that where it primarily targets the inflation rate but does try and correct for past deviations from target, which is essentially the same as targeting the price level just a little bit as well.  

This is does not represent a major change to the way the Fed manages monetary policy, but it is an incremental improvement and encourages the Fed to engage in more stimulative policy in the immediate aftermath of a recession, which could help speed up the recovery.  For now though, only time will tell if this change helps the economy, since it only makes a difference once inflation is about to exceed the 2% target, and we it could be some time before the economy reaches that point.    

What exactly did Jerome Powell announce in his speech?

Today, Jerome Powell made a big speech announcing an update to the overall framework for monetary policy that the Fed uses to set interest rates.  Here are the highlights.

We continue to believe that specifying a numerical goal for employment is unwise, because the maximum level of employment is not directly measurable and changes over time for reasons unrelated to monetary policy.  - Jerome Powell, speech at Jackson Hole, 8/27/20

This is Powell basically admitting the Fed has no idea what unemployment rate starts causing inflation, which has been true for a very long time. 

In addition, we have not changed our view that a longer-run inflation rate of 2 percent is most consistent with our mandate to promote both maximum employment and price stability.

This means the Fed is not raising their target up from 2% to say 3% or 4%.  I believe the inflation target should remain at 2% and not be raised any higher, which I talk about at greater length in my own policy memo.

Our new statement explicitly acknowledges the challenges posed by the proximity of interest rates to the effective lower bound. By reducing our scope to support the economy by cutting interest rates, the lower bound increases downward risks to employment and inflation. To counter these risks, we are prepared to use our full range of tools to support the economy.

Here Powell admits that sometimes interest rates will hit zero and that the Fed will be forced to engage in extraordinary measures, and the Fed is currently very willing to do exactly that.

With regard to the employment side of our mandate, our revised statement emphasizes that maximum employment is a broad-based and inclusive goal.

This appears to be a shout out to the people advocating to add racial equality explicitly to the Fed mandate.  I think this strikes a nice balance, where it acknowledges the need to consider these impacts more explicitly, without elevating it to a place equal to the legislatively mandated goals of price stability and maximum employment.

In addition, our revised statement says that our policy decision will be informed by our "assessments of the shortfalls of employment from its maximum level" rather than by "deviations from its maximum level" as in our previous statement. This change may appear subtle, but it reflects our view that a robust job market can be sustained without causing an outbreak of inflation.

Powell, in this addition to the framework, appears to be admitting that the Fed will not start trying to reverse an economic expansion simply because unemployment is getting too low and will wait until inflation starts going up until taking those steps. 

To prevent this outcome and the adverse dynamics that could ensue, our new statement indicates that we will seek to achieve inflation that averages 2 percent over time. Therefore, following periods when inflation has been running below 2 percent, appropriate monetary policy will likely aim to achieve inflation moderately above 2 percent for some time.

This is also a useful move by Powell, where there is a debate about whether to target the inflation rate or the price level (a topic I address in one of my very first blog posts and in a short paper describing a potential topic for future research).  I say this is basically a false choice (based on insights from the engineering department), where the Fed should primarily target the inflation rate but adjust that target by how much it has missed that target in the past, which means they would be targeting the price level to some degree as well. This is pretty much exactly what Powell announced, so I think this is a good idea as well, and even if it is a modest reform, it does represent an improvement on current policy.

In seeking to achieve inflation that averages 2 percent over time, we are not tying ourselves to a particular mathematical formula that defines the average. Thus, our approach could be viewed as a flexible form of average inflation targeting.

This is Powell admitting that strict mathematical formulas are not going to be used to determine monetary policy at the Fed.  I think this is broadly the right way to go, but I do find formulas useful for guiding policy, even if they should not be adhered to when other factors not in the formula substantially impact the desired course of action.  I will admit that in another policy memo, I suggest the ECB should consider using a QE vs inflation schedule to guide policy in Europe, but do advise that this formula should be revised often as economic conditions adapt over time, so this builds some discretion into the process as well when the need arises. 

In general though, I am thrilled with this revision to the framework for monetary policy used by the Fed.  It avoids increasing the 2% target, but does target an average inflation rate of 2% over time, and leaves it up to the discretion of the policymakers at the Federal Reserve to determine how exactly to do that.  They make other smaller revisions that I think move policy in the right direction and realistically could not have hoped for anything better.  This is just one more example of how Jerome Powell is doing a superb job managing the Federal Reserve in very difficult times. 

Four year lag in policy impacts

In general, my rule of thumb when judging how well a President has governed is to wait 4 years and see what happened in the Presidential term immediately following the one being judged.  By this measure, George W. Bush's second term was a governing disaster because Obama had a difficult time in his first term dealing with the impacts of the housing bubble.  Obama's first term was a governing success because he had a generally easy time dealing with the economy in his second term, even if ISIS now became a threat after Obama withdrew from Iraq in his first term.  Similarly, Obama's second term was largely a governing success because Trump had an easy time for most of his first term, but now it's only taken a 3 year lag for Trump's governing problems to take hold or perhaps more accurately the mistakes of his pandemic response were a little unusual since it only took a very short time to be realized.

You can even go back further and say George W. Bush had a difficult second term because of the mistakes he made in the first term, mostly Iraq.  Whatever troubles George W. Bush had in his first term can be traced to the tech bubble in Clinton's second term and Clinton's inability to deal with Osama bin Laden (though he did try much harder than George W. Bush did before 9/11).  Clinton's second term was largely a success because of the efforts Clinton made in his first term to fix the economy and reduce the deficit.  Now, this rule of thumb is not perfect and the evaluations are a bit subjective, but in general it does take a few years for policy mistakes to rear their ugly head, so you do have to wait a little bit until you really know if someone has governed well or not, so overall I think it works pretty well.