This time its different (part 1)

A while ago, there were a few news stories about how we broke the record for the longest period of continuous economic expansion in US history. I remember hearing the same thing in high school, how, in the late 90’s to early 2000’s, we also broke the longest period of economic expansion.

I thought it was weird that I had lived through two of the longest economic expansions in history; turns out that, in my relatively young life, I’ve also lived through the 4th and 6th longest expansions, beginning in 1981 (shortly before I was born), and 2001, respectively.

Longest Periods of expansion

Longest periods of economic expansion, date refers to start of expansion

So is this just an anomaly? Between 1857 and 1983 (when I was born), we saw a recession about every 4 years, since 1983, its been a recession about every 13. Is something truly different? While I’m not a macro-economist, there are a few thoughts I had about what may be causing this.

Explanation #1: the Fed got smarter.

This explanation is that essentially the Fed, after blundering through monetary policy for the first 65 years of existence, figured things out and are now just better.

Before 1913 (when the Fed was created), there was a recession once every 3.77 years. Between its creation and when Paul Volcker became chairman in 1979, we averaged a recession once every 5 years. Since 1979, we have had only 5 recessions, or once every 8 years.

Vockler Effect

Since Paul Volcker became Fed chair, recessions have become less frequent

There certainly may be something to this, but its the explanation I like the least; both in the sense of explanatory power, and in the sense that I hope its not true. First, it just doesn’t feel right. My memory may be hazy and its hard to compare across periods, but it just doesn’t feel like there were a lot of periods recently where we faced a potential recession, and the fed stepped in an fixed things. If it is true, I think its the worst circumstance, there’s no reason that the Fed will continue to be smart. Sure, they may have access to more history and better theories, but no guarantee that all future chairs will adhere to those lessons.

Explanation #2: the Fed’s job got easier.

Inflation has been generally lower in the recent history. Prior to our arbitrary cutoff of 1983, inflation averaged 2.7% per year, between 1916 and 1982, it averaged 3.6% per year.

Yearly Inflation

Of more importance is its stability, since I was born, inflation was no higher than 5.4% or lower than -0.4%; compared to a range of 18% to -10.5%; the yearly standard deviation of inflation was 5.9% before 1983, but only 1.2% since.  To bore you with one more chart, when examining inflation by decades, we can things getting much more stable (lower highs, higher lows, and a general decrease in volatility of inflation).

Inflation volatility

The implications for the Fed (and therefore the economy) are clear; the Fed’s dual mandate is to fight inflation while maximizing employment. The Fed cuts rates to spur economic growth at the expense of inflation or it hikes rates to fight inflation and hurting growth. If the fed no longer has to worry about inflation, then its free to keep rates low to spur growth.

So why is inflation so low? One reason I think is that demand curves have become less kinked. Imagine, in the 1970s, you have two items your selling, one which you keep running out of, and one that doesn’t sell fast enough. Naive economic reasoning states that you should raise prices on the first and cut prices on the second (as people will buy less of the first, but that’s fine as you’re already selling out of it, and buy more of the second). But in reality, the price hike will more be effective than the price cut? Why? Because you need to advertise the price cut. People won’t magically know about the cheaper prices; if you just change the sticker price you may get more people to buy, say if they’re already in the store, but in general you’ll have to pay more to advertise. On the other hand, raising prices will have the immediate effect, people will immediately see the price as they shop, (and if they don’t for some reason, all the better! Or you may have some customers switch stores because you raised prices, but you’ll probably get one final shopping trip where they’ll pay the higher prices). So while the law of demand still held, it is “kinked,” lowering prices isn’t as effective as raising them.

This means that on average, every company is going to be more likely to raise prices than lower them; which naturally leads to inflation.

The internet changed this though, its much easier to check prices over the internet, cheaper to target advertisements to specific groups of customers (or even specific customers), or more effective to check prices online before going shopping (or even while shopping!). Which means that demand curves got less “kinky” due to the internet; which stops the price ratchet; now cutting prices is more likely to lead to higher sales than it was previously.

I think there’s something to this explanation, although it doesn’t entirely fit the data; inflation fell rapidly in the early 80’s, and gradually since then. We should probably expect the internet to have had no effect before the mid 90’s, small effects in the late 90’s, and then ramp up since then. Looking at the 5 year trailing moving average of annual inflation, we see essentially no fall between 96 and 2008; although a real decline since then.

Other factors which may have led to the fall in inflation include increased globalization of supply chains, decreased relative reliance on oil, and shift to value retailers such as WalMart.

Explanation # 3: Lean manufacturing.

One of the more interesting things I learned in business school was about supply chain management. To condense several weeks of classes into a paragraph, the gist of how to run a supply chain is thus:

You want to have all your machines/people doing stuff as much as possible. Which means you don’t want bottlenecks (one part of your assembly line causing a backup). The cause of this is variability: if tasks take longer (or shorter) than expected, you end up with some machines not being used, (if it takes longer than expected, those things in front of the machine in question are underutilized, if it takes shorter, the machine itself is under utilized).

The naive solution to this is to flood your channel with inventory. At each station, have a number of extra works in progress ready to move onto the next machine; if one machine slows down, those ahead of it can work through the backlog; if it speeds up, it can work on its own backlog.

The cost of this is twofold. First, more of the firms capital is tied up in inventory. This is small but significant. Second, it makes it harder to change production. In effect, having such a huge backlog means that any decisions about how much to produce take a longer time to implement. If you want to cut production, you have to do it ahead of time, as you have a lot of works in progress. If you’re worried that something won’t sell, you may have more of an incentive to cut production sooner, if you wait until its too late then you may end up sitting on inventory you can’t sell. There are similarities with retailing (which, is nothing but the last link in every value chain), if you don’t know how much you’ll sell (and what you’re selling won’t rot), you flood the channel with inventory.

Thus, we may see companies reduce production to perceived future drops in demand. In essence, the harder it is to steer a supply chain. The more any individual company has an incentive to reduce overall production, (or just change production levels for any other reason). This can lead to predictions of recessions becoming self fulfilling; companies reduce production in fear of a recession, and if enough of them do so, it causes that recession.

In postwar Japan, Toyota designed a new way of running their assembly lines. To be very brief, it changed the “direction” of the supply chain; instead of running inventory from the beginning of the chain to the end, it ran requests for work from the end of the chain to the beginning. This is often known as “lean manufacturing” or the “Toyota Production System,” or “just in time inventory.” Overall, it is more efficient generally, and much more efficient with regards to inventory. As the technique spread from Japan to the rest of the world, companies became not just more efficient, but more nimble, you can react to changes in demand much quicker if your supply chains aren’t filled with works in progress.

Overall, more nimble reactions means fewer overreactions which leads to more economic stability. FRED only goes back to 1992 for this stat; but we can see a general decline in inventory levels at least throughout the 90s and 00’s, although it has risen since 2015.

Inventory to sales

Explanation #4: Better forecasting.

We’ve gotten much much better at forecasting than we were in the past. Autoregressive Integrated Moving Average models were introduced to world by Box and Jenkins in 1970; since then the number of people trained in them and the number of computers able to run them has only gone up.

But consider even simple forecasting. If you were managing a department store in the 1950’s, and trying to figure out how much of some product to buy for the Christmas shopping season, how would you do it? What tools would you have available? Well, my guess is you’d probably look to what you did last year and do much the same. Finding trends, even simple trends like “every year for the past 5 years the sales of product x increased 2%” can be very difficult to calculate by hand, especially when you have hundreds of goods and anything less than ideal record keeping. It was difficult, time consuming, and error prone.

The supermarket scanner combined with even simple spreadsheets made it much much easier to how much of and when things sold. Even a simple rule such as “buy 1% more of each good than we did last year” was now easy to do.

As forecasting became better, there was less unsold inventory (and fewer missed sales opportunities), which means less variability in economic activity, leading to fewer recessions.


I’m sure there are things I missed which contribute to the frequency of recessions, and I think the above reasons intermix.  The effects of better forecasting are magnified when supply chains can be changed more rapidly. More efficient production leads to lower costs and less inflation.  An intelligent Federal Reserve becomes better if its job is easier.  There may be other explanations, as an economy grows more complex, it may become more diversified and the risk of any one sector plunging us into recession is reduced.

On Universal Basic Income

As democratic presidential candidate Andrew Yang is getting at least a little attention, and his major idea is universal basic income, I thought I would write a little about it.

Note that there are plenty of ways to implement UBI, and it would be difficult to address them all, so I will specifically addressing UBI as described by Andrew Yang on his website.

The basic idea of UBI is that every adult citizen will receive a certain income from the government with no strings attached (except one important one, which I’ll address in a second), in this particular case $12,000 per year.

The biggest benefit of UBI is that it doesn’t discourage earning additional income.  Current welfare is often tied to (and scaled back with) income. So for instance, people who live in public housing frequently pay a mandated 30% of their income on rent, giving them an effective 30% marginal tax rate from housing benefits alone.  If they earn 1000 more a year, their rent will go up 300, so even before other taxes or reduction in benefits, they only get to keep an additional 700 per 1000 of income. If they receive other benefits (food stamps, etc), they could easily have an effective marginal tax rate of over 50% (compared to a federal high of 37%).

UBI would not have this flaw as anyone working to get ahead would keep the entirety of the benefit (less taxes) regardless of their additional work. So a low income person who works overtime would keep their UBI plus their wages, they aren’t being discouraged from working in the same way that other welfare benefits do.

The other benefits that UBI promises aren’t as great.  UBI will potentially reduce bureaucracy, as there won’t be needs to apply stringent criteria for benefits, for instance, to claim disability, you have to demonstrate that you’re disabled. This may also reduce fraud, (harder to have fraud if everyone is entitled).

There are other claims about UBI, that it will increase entrepreneurship (as people may be more likely to take risks knowing they have a backup), would stimulate the economy, and provide various other benefits.

There’s also an argument that we will need an UBI as advances in technology will make workers obsolete, leading to gigantic increases in unemployment.

However, there are significant problems with UBI, which I’ll go into in detail:

1. As constructed, it will hurt the neediest.

I want to differentiate here between needy and low income, as although they are similar concepts with lots of overlap, they are not the same.  A needy household is one with low income and high expenses, typically either because of the need to support children, or as a result of medical issues. A low income household is (unsurprisingly) one without much income. You can be low income and not be terribly needy; for instance college students and singles without obligations may have low incomes, but I would not consider them “needy.”

The single largest method of funding the UBI would be a Value Added Tax (or VAT), which imposes a tax on the value added to each stage of a products production. The VAT is more or less passed on to the end consumer of a good. In this way, a VAT is mostly flat, in that the rich and the poor will pay the same portion of their income on it (or even potentially regressive, assuming that the rich save more than the poor).

The benefits of the UBI will go to each adult, but people will be prohibited from taking other welfare. This means that the neediest among us would get less net benefit from the UBI, while paying an equal share of their income. This will lead to small or potentially negative effects for the very poor.

Take an example; lets assume there is a single parent household that currently gets 8,000 in public assistance, and makes another 6,000 through part time work. Lets also assume that the increase in VAT will be a straight 10% increase in cost of all goods and services. Before, the household would get 14000 in income. Under the UBI, their income would increase to 18,000 (12,000 in UBI and 6,000 in wages). The general price level would increase though due to the VAT, leaving them with the purchasing power of 16,360 (paying 16,360 for goods and 10 percent of that (1,636) for increase in price due to VAT, totaling 18,000). So the needy, single parent household is better off by only 2,360 per year, less than 20% of the proposed benefit.

Compare that to a non-needy but low income person, for instance a college student. Making the same 6,000 from part time work, their new income is also 18,000; or 16360 after paying for the VAT. But the net change is 10,360; 4 times the net benefit the needy family gets.

Its also easy to imagine people who would get no benefit from the UBI; for instance those on Medicaid with chronic medical conditions. Also, I couldn’t find whether Yang considers social security “welfare” or not. Given that the average social security payout is 1,400 a month, retirees would not benefits from UBI; but again would help pay for it every time they make a purchase.

Going back to the same, theoretical household with 6,000 in income, but this time with 12,000 in benefits, they would see a net decrease of 1,630 in their purchasing power.

There are other ways the UBI can sap the neediest of buying power. For instance, it could lead to rise in rent for low income housing; further deteriorating the purchasing power of low income individuals.

2. It will discourage work via two different mechanisms while increasing the government budget by 50%.

First, as it will be associated with a large tax increase; it will discourage workers (especially higher end workers) from working, as their labor will be rewarded less. High income (and high productivity) workers would, on the margin, retire earlier, work fewer hours, and high earning foreign workers would be less likely to emigrate to the US.

Second, it will lead some people to stop working as they no longer need the money. To some extent, this is the point of the program. Yang talks about people free to be caretakers for their loved ones, for instance. And while this will be good for society, it will be bad for GDP (or at least the taxable portion of GDP).

Assuming there are 200 million adults qualifying for UBI, (my quick estimate of 252 million adults, times 93% who are citizens, less 15% who are on medicaid), the cost would be about $2.4 trillion per year, compared to a current federal budget of about $4 trillion. (incidentally this is where I get the 10% VAT number from; with GDP about 20 trillion, the program costing about 2 trillion, gives us about 10% of the economy we would need to raise in increased taxes).

So essentially, we will be drastically increasing the amount of money we need to raise while reducing the underlying tax base.

Yang argues that a UBI would stimulate the economy. This is based on a study from the Roosevelt institute, showing that a UBI would increase GDP. There are two problems with citing this. First, Yang lies about the contents of the study. Second, the study itself is highly suspect.

First, dealing with the lie. On the “about UBI” page, under the “What impact would Universal Basic Income have on the economy?” Yang says that a UBI would “permanently grow the economy by 12.56-13.10 percent” The source is clearly table 3 of the associated study, looking at scenarios 3 or (12.56 % GDP growth) and 9 (13.10 percent GDP growth). Yet those scenarios look at financing the UBI via deficit spending (essentially paying for the benefits by increasing the national debt, not by raising taxes). Yang proposes paying for the UBI by increasing taxes, which would be scenarios 6 or 12 in the table, or 0% to 2.6%, much more modest than the 12% cited by Yang. This isn’t a simple mistake regarding the signature proposal for the Yang campaign, this is telling a lie, hoping that most readers won’t read through the study, or pay enough attention to decipher the somewhat confusing tables.

Second, the study itself is, I hesitate to say anything disparaging here, well lets just say I think the study shouldn’t be considered the final word on things.

There are essentially two assumptions the study makes. First, that we are far enough away from full employment that we don’t need to consider it. Second, that people will not change behavior due to UBI, either because of higher taxes, impacts of borrowing (ie, government borrowing crowding out private investment) or people becoming less likely to work as a result of receiving benefits.

Addressing the first assumption, under this assumption, there is no way that government borrowing will not increase welfare, no matter the amount. Economic theory states that there is some capacity, based on technology, physical capital, land and human capital to how much we can produce, this is of course true; to continue to grow, we must invest in new factories, new technology, and in ourselves. We may, at any given time, be producing less than that. This is particularly true in recessions; we produce (and consume) less, not because we have seen a loss in productive capacity, but because we don’t produce everything we can. Since about the 1920’s, the dominant macro-economic theory has been that government spending, even government spending on otherwise useless things, can help us reach this potential GDP. Government spending will “inject” money into the economy, causing people to spend more, increasing the economy, causing people to spend more, etc.

Of course this can only go so far; once we’re at the capacity of the economy, we cannot increase GDP by otherwise useless spending. We can only increase GDP by increasing our ability to produce, ie building infrastructure, machinery, technology, or by teaching people new skills. The study simply assumes that we are not at this point, and will not approach it, even with an extra 2 trillion in government spending. Of course, given these assumptions, we could increase government spending by 20 trillion, and the economy would double. Simply put, the study, as constructed, will never find out how spending is too much. Therefore, we cannot trust the study when it says that the proposed plan isn’t too much.

The other assumption I’ve already touched on, namely that we won’t see any discouragement of work either through higher taxes, higher borrowing, or less need to work. I think this is also a very suspect assumption, especially given the sums of money we’re talking about.

Of course, under these assumptions, neither the UBI or any government would ever be seen as anything other than good. Building pyramids, or a giant ditch-digging and refilling program when unemployment is at 2% would both show zero real costs.

3. It will create long term inequality, and create a generational experience gap.

Yang and other supporters present the UBI as a way for people to spend more time as caretakers, artists, or students. While there are certainly a number of people who would use the UBI to pursue these endeavors, there would realistically be quite a few who use the UBI to stay home watching Netflix, playing video games, or otherwise dropping out of society. I can easily imagine somebody doing this for through their mid 20’s, then deciding that they want something else, and having a very difficult time getting out of that cycle. The skill gap caused by this would create a two tiered society; exacerbating inequality and creating a permanent underclass.

4. It isn’t needed, because robots aren’t taking our jobs.

The refrain from UBI advocates that we need it because people are losing jobs to automation, and technology will soon lead to people being unable to find work. The classic example is self checkout kiosks at grocery stores; technology allows the grocery store to run with fewer employees.

This of course nothing new. Fears about automation replacing workers have existed since at least the 1810s, when the Luddites smashed looms in order to preserve their jobs as weavers. The industrial revolution, assembly lines, and computers all were supposed to destroy jobs. Of course they did; but every time they created more jobs than they destroyed.

This is no different than today. Unemployment is at 3.6%, the lowest its been since the late 60s.

People may counter that unemployment isn’t a complete figure, that it doesn’t include “the discouraged worker” phenomenon. Since unemployment measures the ratio of people who are looking to work over those who are working or looking for work, it doesn’t include people who aren’t looking for work. There are various reasons this is appropriate, for instance retirees and students aren’t included as unemployed, nor should they be. But people who have just given up looking for work aren’t included either, meaning that the unemployment rate can be misleading.

To counteract this, we can look at the labor force participation rate; which shows the percent of civilians who are looking for work or working:

There’s been a small decrease in labor force participation since the late 90’s; but its much higher than it has been from the 50’s to the 70’s.

With unemployment near 40 year lows, and labor force participation very high, its safe to say that any displacement of workers by robots hasn’t happened, just as the assembly line or automated loom or computers haven’t permanently destroyed jobs.

Of course, as unlikely as it may sound, this time may be different. We may once and for all see a permanent displacement of workers. In that case, a UBI may be appropriate; if we are ever in a position where 20% of the workforce is permanently unemployable; the arguments for a UBI start looking much better. It just makes no sense to me why we need to start a UBI now in preparation for that. If we get massive amounts of automation, we should see a massive increase in productivity, GDP and overall incomes. In short, when we need a UBI, the question of how to pay for it will answer itself.

But implementing a UBI now because of something that might happen in the future is nonsense. It is a bit like leasing a Mercedes now because we might need a car at some point in the future.


The most compelling reason for UBI is that, for those families who need assistance, we can give them aid without discouraging work. This is a real benefit; but there are ways to do this which cost much less than 2 trillion. My suggestion would be to guarantee benefits for 5 year periods. So for instance, if somebody qualifies for food stamps, they cannot have that revoked for five years, no matter how high their income gets in that period.  Regardless, the problem of discouraging work among welfare recipients, while large, is not a $2 trillion problem.

Overall, I find the arguments for UBI to be lacking and the benefits of UBI greatly exceeded by its costs. There is simply no reason to create a new entitlement spending twice the size of social security.

Oscar Prediction Results

So how’d I do on my Oscar predictions? Lets see?

For the short subjects, I picked solely based on name, which netted me 1 of two; winning on Period. End of Sentence, missing out on Skin.

Doc Short Subject – Period. End of Sentence
Live Action Short – Skin  (running total) 1/2

Animated Short- the rule is never go against Disney/Pixar, which worked well for me here.

Animated Short – Bao – 2/3

Documentary Feature – I figured that the academy would go political, but they did not. Free Solo won.

Doc Feature – Free Solo – 2/4

Neither Makeup & Hair nor Costume Design went with British period dramas, so I’m 0/2 there. Should have trusted the “Black Panther will win every technical category” rule, but I did not.

Makeup & Hair – Vice
Costume Design – Black Panther – 2/6

Visual Effects didn’t go with Avengers; every other category I missed I got wrong; I guess that not including BP here messed up the rules?

Visual Effects – First Man – 2/7

Shallow was definitely going to win. This is my most “buzzy” prediction in that I was merely repeating what I heard, but nevertheless, I got it right.

Original Song – Shallow – 3/8

Looking back, Mary Poppins was the worst pick I made. I could defend any other, but Poppins was just, what was I thinking? Anyway, Black Panther won another.

Original Score – Black Panther – 3/9

My technicals didn’t include Score or costumes and they should have, but at least I picked BP for Production Design. Missed on Bohemian Rhapsody.

Production Design – Black Panther
Sound Mixing – Bohemian Rhapsody – 4/11
Sound Editing I went against the rule which said Black Panther, and picked A Quiet Place. Neither worked, it was Bohemian Rhapsody, which I guess gets every mixing/editing award? No idea what happened there.
Sound Editing – Bohemian Rhapsody – 4/12

Film Editing, I said whoever won this wins Best Picture, I was wrong on that, and my pic (of the Favourite).

Film Editing – Bohemian Rhapsody – 4/13

Foreign Language film I nailed though, I knew it wasn’t winning Best Picture, which made it a lock for FL Pic.

Foreign Language – Roma – 5/14

Missed on Cinematography (missed on my backup as well)

Cinematography – Roma – 5/15

Missed on the screenplays; although I did get that they were giving Spike Lee a lifetime achievement award, I just said director instead of screenplay. I picked The Favourite and If Beale Street Could Talk; missed on both.

Original Screenplay – Green Book
Adapted Screenplay – BlacKkKlansman – 5/17

Broke the rule and went with Spiderman over Incredibles 2, and very happy I did.

Animated Feature – Spiderman – 6/18

Missed on Director – nothing much to say there.

Director – Alfonso Cuaron (Roma) – 6/19

Got Supporting Actress, very happy with my logic there.

Supporting ActressRegina King  – 7/20

Didn’t have a good process for picking supporting actor, so I missed.

Supporting Actor – Mahershala Ali – 7/21

My biggest mistake was thinking that a single movie would sweep various categories (including Best Pic), and while that happened, I chose The Favourite instead of Green Book. Despite that I missed the one Oscar they won. Drat! I guess I thought Lady Gaga was more of a novelty than she really was.

Actress – Olivia Colman – 7/22

Best Actor I crushed though, the academy loves iconic musician portrayals.

Actor – Rami Malek – 8/23

Again, I picked The Favourite as the winner, should have picked Green Book. I thought that Green Book was just too stereotypical, to milquetoast, too “lets fight racism but not actually challenge any of our viewers” to win. Which was dumb, that is the type of movie that Hollywood loves the most! Anyway, I totaled 8 of 24.

Picture – Green Book – 8/24

So I picked one third correctly; which I think is somewhat bad but better than chance. Anyway, lets formalize the rules:

1: There will be one film that sweeps or nearly sweeps the technicals, and it will usually be obvious.

2: There will be another which wins all the prestige categories (but not necessarily acting Oscars). This movie will not be the same as the movie in #1, except in extreme circumstances.

3: Hierarchy for animated (short/feature) is Pixar > Disney > Others.

4: There are certain performances that the Academy loves, but usually biopics of famous people.

5: Look for potential to split votes and don’t go with either in acting categories.
New rules I’d add:

6: Best Score is just another technical, go with that.

7: Makeup hierarchy is probably “Makes Somebody look like somebody else” > British Costume Drama > Science fiction

8: Above all, ask not “which film was best” but “which film makes the academy feel good about itself” that’s the key.
Maybe a rule I should add is “if a movie from #1/#2 isn’t nominated in a given category, don’t pick the one most like it remaining; pick another angle.”
Anyway, that’s my Oscar results post!

Electoral Tilt

How unusual was the 2016 election?

Trump of course, won the 2016 election without winning the popular vote, the 4th time that happened. And he didn’t just do that, he seemed to lose the popular vote by a lot (2%), with a fairly large electoral lead (306 to 232); it seemed like Trump had a very big advantage. The last time we had a popular/electoral split (2000), Bush lost by only a half point in the popular vote, while amassing only 271 electoral votes (against the 270 needed to win). Essentially, 2000 was a statistical tie both electorally and popularly.

We can therefore conclude that Trump had a much larger advantage in 2016 than Bush did in 2000 (and of course both had the advantage over Hillary Clinton and Al Gore, respectively). The map was tilted in Trump’s favor. But how big was the tilt compared to average?

There are several ways to measure this, but the one I chose was to look at the tipping point states; that is, line up all the states in order by the largest margin of victory for the winner to the smallest while adding up electoral votes; the state that brings the total to 270 electoral votes (or whatever the required margin was for the election in question) is the tipping point. In 2016 it was Wisconsin. In 2000, it was famously Florida.

Then compare the tipping point margin to the popular margin; the difference is the advantage. For instance, Trump won Wisconsin by .77% while losing the popular vote by 2.09, giving him an advantage of 2.86 percent. Obama in 2012 won Colorado by 5.37 and the popular vote by 3.76, giving him an advantage of 1.61. (so had every state moved toward Romney by 4%, Obama would have lost the popular vote but won the election anyway).

Ok, now that we’ve defined our methodology; exactly how unusual was Trump’s electoral college advantage, in comparison to his popular vote total.

The answer is fairly. He had a fairly unusual advantage. The average spread between 1856 (the first year which featured Democrat and Republican candidates) was 1.83. Well technically, the average spread was 0, since for every candidate with an advantage there was the other candidate with a disadvantage; but the size of the spread for the candidate with an advantage was 1.83. It seems that the size may be declining over time, if I had to guess I’d say that the most likely reason for this is the increase in total number of states; as more states joined the union, the sample sizes get larger, so differences tended to even out.


Electoral Tilt

Electoral Tilts since 1856

Since 1912 (when Arizona entered the Union, giving us 48 states) the average is only 1.67; since 1960 (the first with all 50 states) the average drops down to 1.05. There hasn’t been an advantage as large as Trump’s since Dewey’s 4 point advantage over Truman (which Truman still won). Third party bids also seem to impact the total, in elections where a third party wins at least one state the average advantage is 3.05, it drops to 1.54 when the race is 2 way.

What about year over year changes? The average change between 2 consecutive elections is 2.2; although that drops to 1.7 when the election features an incumbent (here I’m defining incumbent as a winner of a previous presidential election; so Johnson in 64; Truman in 48, etc, are not “incumbents” for this purpose). In the 18 elections featuring an incumbent, half of the advantages got better for the incumbent, half worse. The advantage regressed towards the mean (including regressed across the mean) about 72% of the time regardless of whether the election had an incumbent or not.

Democrats had the advantage 37% of the time, Republicans 63%, although I doubt if those numbers have much meaning; partisan identification has rotated a few times since reconstruction. The overall trend doesn’t show much pattern: the below chart shows the advantage over time (points below the horizontal axis favor republicans, above favor democrats).

Tilt over time

Electoral Tilt has seemed to decline in the past 50ish years

Grant (R) had a huge disadvantage in 1872; Hayes (R) had a huge advantage in 1876. Harding (R) had a giant advantage in 1920; (second largest since 1856); Coolidge (R), Harding’s running mate, faced a large disadvantage in 1924.

So any conclusions? Trump had a large electoral tilt which helped him win the Presidency; he would have lost had he only a normal sized tilt in his favor and the same proportion of the popular vote. Large tilt’s usually don’t persist; about three quarters of the time the tilt regresses towards zero. An incumbent only keeps the electoral tilt half the time (if they start out with it). All of this seems to say that Trump probably can’t count on as favorable a map as he had in 2016, in order to win reelection he will probably need to close at least some of the gap in the popular vote.

Finally, I received most of this data from ; so thank you to JRP1994; I couldn’t have written this post without you.

On the Wealth Tax

Lets talk about the proposed wealth tax. Proposed by Elizabeth Warren, it would put a tax of 2-3% on wealth above $50 million.

How exactly does one determine wealth? Assets less liabilities one would imagine. Seems simple on paper, but there’s a reason why accountancy is a profession, and a high paying one at that. How are asset values determined? For things like bank deposits, bonds and publicly traded stocks, there’s not too much concern. But what about other assets? Real estate we already have assessors, but real estate is kind of easy, there are comparable properties.

So there are two basic ways to value a company (I’m sure there are hundreds, but lets focus on two). First, you take the companies assets, subtract its liabilities, and viola, you get its value. The second, is you discount the expected future cashflows; this is generally considered how companies are valued on the stock market. These give you radically different values. For instance, Microsoft is worth about $47 billion according to the first method (or $83 billion, depending on which assets you include) according to the second method, $823 billion. So a difference of around 10 times.

Now lets say you own a private company- how is this company valued? You can’t compute based on market price as there are no public prices. Performing a discounted cashflow (DCF) analysis is an option, but you get wildly different opinions based on the inputs. Whenever there’s a merger, the advising investment bank will perform a “fairness opinion,” typically using a DCF, to demonstrate that the price is fair and to avoid legal liability. They always get horribly useless numbers, like $8 to $70 dollars per share, (meaning that the merger is fair if the acquiring company pays more than $8 but less than $70 per share).

Maybe the government will standardize the way private companies are valued; Say that a company is worth X times its yearly earnings; and that the owners would have to pay a percent of that. Say X is 10; so a private company with 100 million in earnings would be worth 1 billion, and the tax liability for the owners would be 20 million a year and – hey wait a minute – this seems familiar.

wealth tax scooby doo
So maybe we don’t go with that method, but the fact remains that changing the way the government values private corporations will have enormous impacts on actual economic transactions, companies will probably be much less likely to go public, (or maybe more likely depending on the actual rules). We may have less capital available, or companies may favor debt financing vs equity financing (increasing total risk in the system); people may move to storing their value in unproductive assets (gold, art, antiques, yachts) rather than assets which can be fairly valued. (And how do you value a piece of art? What it was bought for? What similar pieces sold for recently? There are lots of issues with art price indexes which cause biases).

And that’s before we talk about people leaving the country or hiding assets abroad. Sure we may have that already, but think of how much worse it could get. But also; it has a particularly perverse effect. It will make it cheaper for foreigners to own domestic companies than for citizens to.
Which brings us all back to old man corporate income tax. We have a method of taxing wealth in this country; it hits foreigner and citizen alike. It treats private and public companies the exact same; the wealthy already pay it, even if the check is signed by a corporate treasurer. And it is fairly tough to avoid it by moving overseas. And while I’m sure that Senator Warren wouldn’t mind raising the corporate income tax, its not as interesting or “sexy” as a wealth tax, which is one of the reasons she introduced it as part of her presidential campaign. Simple, broad based taxes are almost always less disruptive and more efficient than fancy taxes designed to advance some agenda. If the government needs more money, we should increase already existing taxes, (or move to a proven efficient tax such as Value Added or sales); not create a new tax which brings a load of pitfalls.

Oscar Predictions

So I haven’t written anything in a while, so I figured I’d waste my time and yours with…. Oscar Predictions.  I haven’t been following the buzz at all, and have seen maybe two of the below films, but whatever; shouldn’t hinder me.  Oscars are about meeting expectations, not about film quality.


Best Documentary Short Subject:

“Black Sheep,” Ed Perkins
“End Game,” Rob Epstein, Jeffrey Friedman
“Lifeboat,” Skye Fitzgerald
“A Night at the Garden,” Marshall Curry
“Period. End of Sentence.,” Rayka Zehtabchi

Best Live Action Short Film:

“Detainment,” Vincent Lambe
“Fauve,” Jeremy Comte
“Marguerite,” Marianne Farley
“Mother,” Rodrigo Sorogoyen
“Skin,” Guy Nattiv

No idea!  There are no rules for short films; to get a good idea you probably have to actually watch them.  For absolutely no reason I’m going with “Period. End of Sentence.” and “Mother” because why not?

Animated Short:

“Animal Behaviour,” Alison Snowden, David Fine
“Bao,” Domee Shi
“Late Afternoon,” Louise Bagnall
“One Small Step,” Andrew Chesworth, Bobby Pontillas
“Weekends,” Trevor Jimenez

When in doubt, go with Pixar.  Bao is my pick

Best Documentary Feature:

“Free Solo,” Jimmy Chin, Elizabeth Chai Vasarhelyi
“Hale County This Morning, This Evening,” RaMell Ross
“Minding the Gap,” Bing Liu
“Of Fathers and Sons,” Talal Derki
“RBG,” Betsy West, Julie Cohen

RBG, based solely on the subject matter.

Makeup and Hair:

“Mary Queen of Scots”

Costume Design:

“The Ballad of Buster Scruggs,” Mary Zophres
“Black Panther,” Ruth E. Carter
“The Favourite,” Sandy Powell
“Mary Poppins Returns,” Sandy Powell
“Mary Queen of Scots,” Alexandra Byrne

The rule for Makeup and Costumes is you always go with classy British period dramas.  You could argue for Black Panther, but I’m ignoring my gut and going with the rule; Mary Queen of Scots for Makeup and The Favourite for Costume Design.

Visual Effects:

“Avengers: Infinity War”
“Christopher Robin”
“First Man”
“Ready Player One”
“Solo: A Star Wars Story”

Would be Black Panther, but its not nominated.  Will be Avengers.

Original Song:

“All The Stars” from “Black Panther” by Kendrick Lamar, SZA
“I’ll Fight” from “RBG” by Diane Warren, Jennifer Hudson
“The Place Where Lost Things Go” from “Mary Poppins Returns” by Marc Shaiman, Scott Wittman
“Shallow” from “A Star Is Born” by Lady Gaga, Mark Ronson, Anthony Rossomando, Andrew Wyatt and Benjamin Rice
“When A Cowboy Trades His Spurs For Wings” from “The Ballad of Buster Scruggs” by David Rawlings and Gillian Welch

Going with Shallow.

Original Score:

“BlacKkKlansman,” Terence Blanchard
“Black Panther,” Ludwig Goransson
“If Beale Street Could Talk,” Nicholas Britell
“Isle of Dogs,” Alexandre Desplat
“Mary Poppins Returns,” Marc Shaiman, Scott Wittman

I could argue for anything; but I think I’ll go for Mary Poppins Returns.

Production Design:

“Black Panther,” Hannah Beachler
“First Man,” Nathan Crowley, Kathy Lucas
“The Favourite,” Fiona Crombie, Alice Felton
“Mary Poppins Returns,” John Myhre, Gordon Sim
“Roma,” Eugenio Caballero, Bárbara Enrı́quez

Sound Mixing:

“Black Panther”
“Bohemian Rhapsody”
“First Man”
“A Star Is Born”

There is always one movie which sweeps the technicals, one which sweeps the prestige.  I’m not sure who is sweeping the prestige Oscars, but Black Panther is gonna win most of the technicals its nominated for, including Sound Mixing and Production Design.

Sound Editing:

“Black Panther,” Benjamin A. Burtt, Steve Boeddeker
“Bohemian Rhapsody,” John Warhurst
“First Man,” Ai-Ling Lee, Mildred Iatrou Morgan
“A Quiet Place,” Ethan Van der Ryn, Erik Aadahl
“Roma,” Sergio Diaz, Skip Lievsay

The rules say to go with Black Panther, but I have to give it to A Quiet Place, cause the whole movie is about sound.

Film Editing:

“BlacKkKlansman,” Barry Alexander Brown
“Bohemian Rhapsody,” John Ottman
“Green Book,” Patrick J. Don Vito
“The Favourite,” Yorgos Mavropsaridis
“Vice,” Hank Corwin

So my guess is The Favourite, but its only a guess.  But very good odds that whoever wins this wins Best Picture.

Best Foreign Language Film:

“Capernaum” (Lebanon)
“Cold War” (Poland)
“Never Look Away” (Germany)
“Roma” (Mexico)
“Shoplifters” (Japan)

Roma will win here cause it won’t win Best Picture (think Life is Beautiful).


“Cold War,” Lukasz Zal
“The Favourite,” Robbie Ryan
“Never Look Away,” Caleb Deschanel
“Roma,” Alfonso Cuarón
“A Star Is Born,” Matthew Libatique

Could go with A Star is Born, but I’m going with The Favourite.

Original Screenplay:

“The Favourite,” Deborah Davis, Tony McNamara
“First Reformed,” Paul Schrader
“Green Book,” Nick Vallelonga, Brian Currie, Peter Farrelly
“Roma,” Alfonso Cuarón
“Vice,” Adam McKay

You know the drill.  The Favourite.

Adapted Screenplay:

“The Ballad of Buster Scruggs,” Joel Coen , Ethan Coen
“BlacKkKlansman,” Charlie Wachtel, David Rabinowitz, Kevin Willmott, Spike Lee
“Can You Ever Forgive Me?,” Nicole Holofcener and Jeff Whitty
“If Beale Street Could Talk,” Barry Jenkins
“A Star Is Born,” Eric Roth, Bradley Cooper, Will Fetters

Hmm.  I think they’re gonna go with “If Beale Street Could Talk.”  Chance they go with Buster Scruggs.

Animated Feature:

“Incredibles 2,” Brad Bird
“Isle of Dogs,” Wes Anderson
“Mirai,” Mamoru Hosoda
“Ralph Breaks the Internet,” Rich Moore, Phil Johnston
“Spider-Man: Into the Spider-Verse,” Bob Persichetti, Peter Ramsey, Rodney Rothman

The rules say go with Pixar.  But I’m breaking the rules here and going with Spider-Man.


Spike Lee, “BlacKkKlansman”
Pawel Pawlikowski, “Cold War”
Yorgos Lanthimos, “The Favourite”
Alfonso Cuarón, “Roma”
Adam McKay, “Vice”

They’re gonna give Spike Lee what amounts to a lifetime achievement award for BlacKkKlansman.

Supporting Actress:
Amy Adams, “Vice”
Marina de Tavira, “Roma”
Regina King, “If Beale Street Could Talk”
Emma Stone, “The Favourite”
Rachel Weisz, “The Favourite”

Normally it would go to Emma Stone, but no-one has won a best supporting actor/actress award if another person is nominated from the same movie since Hattie McDaniel beat Olivia de Havilland for Gone with the Wind; (OK, I totally just made that up so its probably not true), but you will get vote splitting.  So I’m going with Regina King.  Also, I hope that Olivia de Havilland presents an award.

Supporting Actor:

Mahershala Ali, “Green Book”
Adam Driver, “BlacKkKlansman”
Sam Elliott, “A Star Is Born”
Richard E. Grant, “Can You Ever Forgive Me?”
Sam Rockwell, “Vice”

I think they’re gonna give it to Adam Driver, cause that seems like something they would do.

Lead Actress:

Yalitza Aparicio, “Roma”
Glenn Close, “The Wife”
Olivia Colman, “The Favourite”
Lady Gaga, “A Star Is Born”
Melissa McCarthy, “Can You Ever Forgive Me?”

Lady Gaga.  Think like Cher in Moonstruck.

Lead Actor:

Christian Bale, “Vice”
Bradley Cooper, “A Star Is Born”
Willem Dafoe, “At Eternity’s Gate”
Rami Malek, “Bohemian Rhapsody”
Viggo Mortensen, “Green Book”

If there is one thing that Oscar voters have an irrational love for, its biopics, especially about tragic and/or iconic musicians.  And no one is more Iconic than Freddy Mercury.  I’m going with Rami Malek.

Best Picture:

“Black Panther”
“Bohemian Rhapsody”
“The Favourite”
“Green Book”
“A Star Is Born”

If Roma wasn’t a Netflix movie, it’d be the front runner.  But industry types (rightfully) consider it a threat to their livelihoods; so that one is out.  Bohemian Rhapsody and BlacKkKlansman and Vice I’ve heard aren’t actually good films.  The big win for Black Panther was just to be nominated (they haven’t awarded Best Picture to a Box Office behemoth since Return of the King).  That leaves Green Book, A Star is Born, and the Favourite.  Normally, one movie becomes the favorite and the voters vote for them.  Studios do all sorts of things to jockey for position in the run-up, but we’re about to see if just flat our declaring yourself the favourite has enough priming effect to get your movie into the lead spot.  I think it will.  The Favourite for the win.

Wel (un) fare

Let’s say that you are creating a tax policy for a new society, and you come to the question of corporate taxes. Which of the following do you choose:

1. You tax more profitable companies at a higher rate (so very profitable companies pay 35% of their income, less profitable companies pay 30%, and so on), based on their margins (or if you prefer, return on equity or return on assets or some other preferred measure).

2. The opposite of 1, You tax less profitable companies at a higher rate based on margins (or RoE or whatever).

3. You say that corporate profit is too complicated to compare rich companies vs poor companies, and instead charge a flat corporate tax on net profits.
3 is pretty much where America is right now, and I would argue that it’s probably the best outcome.  You can of course make a case for 1 if you want, (although I think it would be way to complicated to actually pull off intelligently).  There is no real reason for choice 2, either from a fairness or an efficiency standard.

Let’s pretend that you’re a 19th century industrialist, you run a steel mill or something. For the sake of this example, let’s say that the only thing you care about is making the most money possible, you pay your worker’s as little as you can and are always looking to save money, particularly on labor.

Now lets say the government is deciding whether to enact some form of a welfare program; details aren’t really important, except that you expect most of your workers to qualify whether or not they remain employed with you. Do you, purely from a selfish perspective, want the government to enact a welfare program?

There are two competing schools of thought.

1: Yes, you do, because government welfare means that you can pay your workers even less than before.

2. No, you don’t, because welfare means that: a) your employees will have more bargaining power (they can hold out longer, have a better alternative to negotiated agreement, can take a longer time to search for a better paying job); and b) there will probably be some contraction in the total labor supply as some fraction of employees choose not to work (or choose to work fewer hours), the contraction in labor supply leads to a higher price of labor (ie wages).

1 really doesn’t make sense except under certain extreme scenarios, for instance if people are working for literal subsistence wages and a reduction in income would literally kill them. And even then, only if the welfare program is small enough, once it provides enough benefit, the effects from item 2 overwhelm the effects from item 1. In general, I am confident in saying that social welfare programs increase the cost of labor for big companies.


Lets say you go into a Walmart, and buy a iPhone. Who does the cashier who checks you out work for? Who do they provide value for? Well, the obvious answer is Walmart, in that Walmart chooses whether to hire them, what wages to pay, writes their paycheck and makes them wear a Wal-Mart specific uniform. But in some small sense, they also work for Apple, in that there is some amount of value that they provide to Apple. The Walmart cashier is the end-point of ten thousand different value chains; in some way Apple (and General Mills, and Reebok, and Proctor & Gamble, and virtually every other company) all employ that cashier.


Bernie Sanders has introduced a bill (Stop Bad Employers by Zeroing Out Subsidies), or “Stop BEZOS” act, which seeks to charge low-wage employers for the total amount of public assistance that their employees receive. (See here).  So, for instance, if a Walmart worker gets 2,000 in public assistance, they would be taxed that full amount. This has been picked up by, of all people, Tucker Carlson, (See here).

I think this is an extremely bad bill, for three major reasons.

First, a purely moral/fairness reason. As a society we have decided that people should have some minimum standard of living, and have implemented various programs to pay for them. I strongly believe that we as a society should pay for them, that in effect everybody who can should contribute towards such benefits.

Apple has about 123,000 employees and yearly revenue of 229 billion dollars, or about $1.8 million in revenue per employee. (source, Apple financial statements). Walmart, on the other than, has about 2.3 million employees, and yearly revenue of about 500 billion dollars, or about $217,000 per employee. Apple in turn makes about three times more money ($61 billion in operating income vs $22 billion for Walmart). I’m sure that the average Apple employee makes a lot more than the average Walmart employee, but that has little or nothing to do with the generosity of the employer, and almost everything to do with the type of business they’re in and the difference in skills needed to be a hardware engineer vs a greeter. This bill would in effect place a major tax on the comparatively poorer corporation (Walmart, in one of the few comparisons where Walmart is considered relatively poor) while leaving the comparatively richer corporation (Apple) at a lower tax rate. Why? This goes back to the first question I posed, about tax rates for corporations why should the more profitable company pay less in taxes than the less profitable company? It’s not even that Walmart chooses to use lower-skilled labor, Apple does as well, they just outsource it to China or Foxconn (on the production side) and sellers such as Walmart on the distribution side.

Sure, you can say that Walmart can afford to give employees more, but you can say that so much more about Apple, Google, Facebook, Microsoft etc. The commanding heights of the economy are in information technology and finance. To suppose that retail should bear the burden of welfare is arbitrary and makes little sense. If we are going to support the poorest among us, then I think we should all contribute, (including through personal income taxes), but certainly there is no reason to suppose that Mark Zuckerberg has less means to give than the Walton family.

The second reason is the fact that, as demonstrated in my second example, employers don’t benefit from the welfare at least not in terms of reduced labor costs. In fact it probably raises their costs. Without a reduction in labor costs, there shouldn’t be

The third reason is that this policy would create horrible incentives from just about every angle. With greatly increased labor costs, companies would reduce low-skill employment, either by substituting with capital (more self checkouts, for instance), or just less staffing, as in reducing staffing in off-peak hours, for instance.

It would mean that there is a powerful lobby for reducing government subsidies, there would now be a major rich constituent for reducing all government benefits.

There would be a very bad incentive for employers to avoid giving jobs to the most needy. Some companies will find a way to ensure their employees aren’t needy, dont’ have kids, or are otherwise better off. This bill would make teenagers of wealthy parents a better option for many companies than a single mother of three. Finally, there are truly ghoulish incentives, some employers will insist that employees must not apply for welfare. This will no doubt be very illegal, but it will no doubt also still happen.

There is of course no countervailing incentive; this won’t increase the total amount of welfare available to the poor, it will only shuffle around dollars somehow, while moving more people off paying jobs, increasing their total welfare cost while reducing their total income.

Overall, the bill makes no sense, it is based on bad economics. It supposes that the burden of welfare should be on the employers closest to the employees, instead of to society in general (including to the most profitable companies). It assumes incorrectly that welfare programs decrease the cost of labor for major companies. And it ignores incentives which will lead to fewer low skill jobs and further reduces opportunities for the unfortunate.

What we talk about when we talk about Stock Buybacks

So the other day I offhandedly referenced an article on Facebook that I really didn’t like, and thought I might expand upon my why I think it was so horrible.

The article dealt with stock buybacks, and came out very strongly against them. You can read it here:

If you haven’t read it, it basically goes like this:

1: Companies are using a lot of money to buy back shares
2: This diverts cash from other worthy projects, such as investment in infrastructure, R&D, or paying employees more
3: Companies conduct buybacks as a means of moving their stock price in the short term, but because they’re not investing in their future, they are sacrificing their long term prosperity to artificially manipulate their stock, usually so executives can get bonuses.
4: Therefore, society would be better off if companies were to use that cash in better ways, such as pay dividends instead of buybacks

There are really three issues that the article jumps back and forth in between, sometimes without acknowledging so.

The first issue is to what extent our economy benefits the wealthy and/or the capital owners. This issue is of immense importance, but it is kind of a dead end to argue about.  Simply talking about how you don’t like it is merely complaining, and although it may be an issue worth complaining about, the complaints themselves won’t solve the problem.

The second issue is choosing between the short term and the long term. This is an empirical question, but at first glance, it appears to be sort of moral. After all, we the morality tales of the goose who laid a golden egg, the three little pigs, and the industrious ant vs the lazy grasshopper. All of which have the same general message, don’t get too greedy in the short term; you’ll pay for it in the long run.  But moral connotations notwithstanding, t is strictly a financial issue.  The decision to forgo a dollar today for two dollars tomorrow is not just a question that finance can provide insight into, it’s the question that underpins all of finance, the question finance was invented to address.

The third issue is of dividends vs buybacks. This is a boring, technical distinction; of interest only to finance geeks such as myself, and the CFO’s making the decisions. The differences are arcane, arise from tax law or market structure or weird behavioral equilibrium.  It is easily the least important of the three issues.

The article I referenced, and indeed the debate around buybacks overall, wants to argue point 1 (distribution of wealth in society), yet spends most of the time talking about point 2 (now vs later), mixing in point 3 (dividends vs buybacks). You can see this when it pivots from Tung and Milani’s arguments in favor of dividends: “Issuing cash dividends (regular or special) has a less predictable and manipulative impact on a company’s stock price—and thus is less prone to gaming by executives or activist investors for their own gain,” seamlessly to Senator Warren’s comment: “The surge in corporate buybacks is driving wealth inequality and wage stagnation in our country…”

If you don’t have the background in finance, you’ll probably miss this distinction. Tung and Milani are arguing from a very specific framework, that the act of returning cash to shareholders via buybacks is manipulative of the underlying stock price whereas returning cash via dividends isn’t. (I haven’t read their research so I can’t comment intelligently, only to say that a: I am skeptical of their claim, but also b: the argument itself appears to be withing the realm of possibility). Warren is referencing buybacks, but she is really talking about returning cash to shareholders, and there should be no reason why returning cash to shareholders via dividends or buybacks would affect levels of investment or employee compensation.

What the debate seems to miss is the fact that financial markets, while they may not be the best at allocating resources across people, are incredibly good at allocating resources across time. To believe that buybacks allow companies to prop up stock price in the short term while sacrificing it in the long run is to believe that there are a significant number of asset managers who are fooled by an easy trick. Changes in future expected earnings will effect stock prices today. This also holds true no matter what the time frame that the stock owners have. For instance, imagine you own stock in a gold mining company which you intend to sell next year to fund your retirement. The company today announces that they’ve discovered a rich new mine which they can begin mining in two years. Even though you intend to sell before the new mine opens, you will still profit.  The stock would increase the day of the announcement as people buy today in order to get the future profits, driving up the stock price. This is the whole point of the stock market, to exchange money today for future earnings tomorrow.

If buying shares lowers the expected future earnings per share of a company, it would decrease the stock price. If it increases the future expected earnings per share, it will increase it. (In case you’re interested, the decision should be made buy comparing the rate of return on existing future cash flows divided by the stock price to the rate of return on potential cash flows divided by the cost of the capital project; for instance if a stock yields 10% and the best available project yields 8%, then it makes sense to buy back shares. If instead the best available project yields 12% then it makes sense to fund the project before buying back shares).

To be sure, decision makers (either corporate boards or the asset managers) may have biases. They may not think that available projects are as profitable as they really would turn out to be. But most academics think the opposite is true, that company management is more likely to overestimate the profitability of future investments (via overestimating their own competence, of course); they are overconfident, or they seek growth for the sake of growth as management seeks to build an empire regardless of cost.

Of course, there are instances of companies buying back shares at the expense of the future. Perhaps the most famous is Sears Holding. When Eddie Lampert bought Sears and merged it with K-Mart, he spent massive amounts buying back shares, meanwhile he cut both capital expenditures (to about a quarter of even beleaguered competitor JC Penny’s level); and also slashed advertising. This led to a long decline of Sears, once valued at $19 billion, today Sears is only worth around $200 million. This may seem like the classic story that opponents of buybacks spread; buybacks happened in place of investment; store quality declined, the brand suffered, all of which killed the company.

But this leaves out a key component; the short term outlook from management. Lampert didn’t take over, buy back shares, then exit while leaving somebody else holding the bag. He’s still holding the bag; he’s lost billions (his actual performance is more complicated as there were spin-offs and loans and stuff, but I’m confident that although Lampert didn’t lose the full 99% that the stock went down, he’s certainly lost a hell of a lot). This is a story of misplaced optimism or hubris or plain old stupidity; but its not a story of short termism. He really thought that buying back shares was the right move.

All of this leaves us with how we talk about buybacks. People (specifically the journalists, op ed columnists, and senators) want to be talking about the role that capital plays in society; about how the pie gets divided. But its hard to talk about in any way other than complain, to propose meaningful changes about how society functions to achieve those ends. So we end up talking about adjacent issues, in this case we about how buybacks starve some other useful function of cash, then buttress the argument by throwing in some obscure point from an academic which, if examined closely enough, doesn’t actually apply to the argument. We want to think that there’s some easy solution, that there’s some villain, that if we just removed their ability to manipulate the market, that they’d be left with no choice but to invest in the future, increasing wages and restoring economic balance to America. But alas, if such a policy exists, its certainly not buybacks. The fact of the matter is that many companies simply don’t have useful investments. How many more stores can CVS open before they start to simply siphon customers from existing stores (probably not too many). How much more money can Apple spend on R&D before its funding projects with no hope of completion? We want to blame economic inequality on something, but buybacks simply isn’t it.

Does Whatever a Teenager Can

The greatest work of American art is, in my opinion, The Amazing Spider-Man.

A comic book originally created by Stan Lee and Steve Ditko, it chronicles the adventures of Spider-Man, a teenage boy who, having been bitten by a radioactive spider, develops super-powers, including the proportional strength and speed of a spider, the ability to stick to walls, and a sixth sense for danger. He uses these powers to battle evil villains such as Doctor Octopus, a man with 4 robotic arms, and Electro, who can control electricity.

Of course, you already know this. Spider-Man is the perhaps the most famous superhero, and is as recognizable as Mickey Mouse.  There have been six major films about Spider-Man and at least as many cartoons over the years.

Most super heroes have a very clear single nemesis. Superman has Lex Luther. Batman has the Joker. The X-Men have Magneto. The Fantastic Four have Dr. Doom. Captain America has the Red Skull. Each of these examples provide not only the means for drama for each hero, but also represent the antithesis of the hero. Batman is the personification of law and order, while the Joker represents chaos. The Fantastic Four are primarily about how science can benefit mankind, while Dr. Doom represents technological progress gone awry and used to repress. The conflict between the X-Men and Magneto is about the how outsiders relate to the community, whether to engage in it or rebel against it.

With Spider-Man, there doesn’t seem to be a single villain the way there is in other comics. You could argue for Dr. Octopus (who was in a very large number of issues, especially towards the beginning of the series), the Green Goblin (the central villain in the most famous story line), or Venom (btw, Venom sucks; the introduction of Venom is pretty much when ASM stopped being good). Other candidates include the Kingpin (a crime lord), and the burglar who originally killed his uncle Ben in Spider-Man’s debut (Amazing Fantasy 14, for those who care).

But none of those feel right. Most are fairly generic, there’s no real reason most of them couldn’t be villains in other comic books. In fact, some of them are, the Kingpin is the central antagonist to Daredevil. The reason that none of the villains feel like the central antagonist is that none of them are the central antagonist. They’re temporary obstacles to overcome; they may provide the drama in any given issue, but they represent nothing in the grand scheme of things. Sure, the Green Goblin killed Gwen Stacy, perhaps the pivotal moment in all of the comic. But the Goblin’s role was almost irrelevant; the plot didn’t revolve around the Green Goblin killing Gwen Stacy, it revolved around Gwen Stacy dying. There would have been no real change to the Spider-Man character if Mysterio or the Vulture killed her.

So how can I make the bold claim that The Amazing Spider-Man is the best work of American art if there isn’t any real overarching conflict? Simple, Spider-Man did have a nemesis, one better than any other comic book, and infinitely more relatable and relevant to its readers. The real nemesis of Spider-Man is Peter Parker. The central conflict of Spider-Man isn’t between two people with superpowers fighting for good or evil; the superpower conflict is the superficial conflict. The real conflict is between the goals, dreams and ambitions of Peter Parker, a mild-mannered teenager who loves science, photography, and just seems to want a girlfriend, and the responsibilities that Parker has because he’s the only one who can stop whoever the supervillain of the week is.

Peter Parker placed his responsibilities as Spider-Man above his own ambitions; and as is the superhero cliche he felt he could never let anyone know he was Spider-Man, lest they be used against him (ie, a supervillain kidnap his girlfriend,for instance). He gets these awesome powers, and instead of being liberating they’re confining. His mantra, that with great power comes great responsibility, determines everything that he does. He constantly lets people down in his life (standing up dates, for instance) in order to save innocent bystanders. He cannot quit being Spider-Man because that would allow evil to triumph and cannot even depend on his friends or family for emotional support. The great tragedy of Spider-Man is that the more he tries to serve the world, the further he drives himself away from everyone he cares about.

In a very real way this is what being a teenager is like. The system of emotional support children have is, if not gone, changed. Most teenagers won’t run to their parents with every problem the way most children will; if the typical image of children is a screaming toddler, the typical image of adolescence is the brooding teen bothered by something but unwilling or unable to communicate it. For the first time, teenagers decide to compartmentalize their lives, realizing that some things are “cool,” some aren’t; and will start to mold personas to fit in. in doing so, they don a sort of secret identity, being one person at school or in social events, while another very different person while alone. Teenagers struggle with things like romantic attraction and self worth for the first time. “What do you want to be when you grow up” ceases to be a idle question and starts to be something that needs attention and hard work. And the fears associated with these things are often dealt with alone; I remember how horrifying my future seemed when I was in high school. How it seemed like everything I did had an enormous impact on my future possibilities, and how I felt totally unqualified to make any decisions about it, and how I felt I needed to put on a brave face for my parents so they wouldn’t worry, and how all this came together in an incredibly isolating fashion. I had giant concerns and no way to deal with them except to power through and hope for the best.  And more than anything else, even though virtually all of my peers were going through something similar, even though all the adults in my life went through this themselves, I felt that I had nobody I could talk to.  The burdens I bore I bore mainly alone.

Obviously, the things that typical teenagers deal with are (hopefully) not as life or death as the things Spider-Man deals with, but they don’t seem that way to teenagers. With the benefit of hindsight, my first real crush seems almost laughable, but in no way was it laughable to me, it caused me angst and confusion and feelings of joy and worthlessness, of hope and desperation and confusion wrapped up together.  I recognize now how it was foolish, but it meant the world to me at the time.  Eventually I grew up, I learned how to deal with emotions, and I learned how and when to communicate my feelings.

Eventually, Peter Parker grows up, and the conflicts between himself and his alter ego begin to get resolved. Peter marries Mary Jane, and has somebody who he can reveal his true identity to, someone who knows him for who he is and while useless in a battle against a supervillain, provides support and encouragement. The dividing line between the public perception of Peter and Spidey remains as stark as ever, there will always be personal you and professional you, but in private, Peter and Spidey are at peace.

Perhaps the finest moment in all of Spider-Man, or in all of comics, comes in an otherwise weird period of Spider-Man, the mid 1990’s. Most of the comic revolved around the much maligned clone saga, a goofy story line revolving around whether Spider-Man was a clone or not. Meanwhile, Aunt may is dying. In her final issue, she takes Peter to the top of the Empire State Building and reveals that she knows, that she always knew, that Peter is Spider-Man, and that she is deeply proud of him.

What a wonderful message; that we’re never alone as we think we are; that even if your friends and family can’t fight your battles for you, that they still know that you’re facing struggles, and they’ve always been proud of you.

On Charlottesville

There has been a group of people who have been making arguments for what may be called white nationalism. They are essentially taking the ideology and updating it with philosophical underpinnings. They’re changing the image of the white supremacist from a compound dwelling, tattooed skinhead to a khaki wearing, clean shaven suburbanite. These new white supremacists are better looking, and when they’re not in a mob I would even venture to say that some of them are polite and well mannered.

This doesn’t make them better than the traditional neo-Nazi. It makes them worse. It makes them worse because they can’t hide behind drug problems or abuse or some other excuse. They’re not running to Nazism because they want to rebel, they’re joining white nationalism because they really believe it. The alt-rights aesthetic makeover and philosophical mask make them worse than neo-Nazis because it makes them more like regular Nazis.

In essence, they’re trying to make hate more attractive; they’re trying to fool you.


This past Saturday, a group of white nationalists, neo-Nazis, Klansmen, and their ilk marched on Charlottesville, VA, ostensibly to protest the removal a statue honoring confederate general Robert E Lee. This was a minor goal, if that. Should the statue be torn down tomorrow, they will have accomplished everything they wanted and more.

The name of the march was “Unite the Right.” Since World War 2, Nazi’s have been person non grata in American politics, welcome in no party. While we have by no means been perfect in race relations, we have at least kept naked segregationists out of power for the past five or so decades.

“Unite the Right” aimed to change that. They want a seat at the table; for the alt-right to become a voice within the Republican Party.  David Duke, “former” Klansman and neo-Nazi, of whom former President Bush Sr said “has a long record, an ugly record, of racism and of bigotry,” tweeted to Trump on Saturday that “I would recommend [President Trump] take a good look in the mirror & remember it was White Americans who put you in the presidency, not radical leftists.”  He was expressing just that, a belief that white nationalists like him are part of the coalition that put Trump in the White House.

On Saturday, Trump (who had campaigned partially on the idea that we need to name Radical Islamic Terror as such to defeat it), denounced these protesters in the gentlest possible terms; condemning not just the protesters but the counter-protesters as well, writing “We condemn in the strongest possible terms this egregious display of hatred, bigotry and violence on many sides, on many sides.”

As an aside, for anyone who thinks this moral equivalence is OK, I have three points to make about that:

1. This country has a troubled history of racism; its not an exaggeration to say that a majority of our social problems even today are due to the legacy of slavery, segregation and bigotry, and the lingering insidious racism that millions of Americans are forced to deal with every day. The Klansmen, Nazis and white nationalists represent and seek to reintroduce this evil, the anti-fascists do not.

2. The white nationalists drove a car into a crowd, injuring 19 and killing one. This wasn’t an accidental killing, but rather a white nationalist decided to use his car a weapon. I don’t support the violence of the counter-protesters, but to equate it with the violence used by the white nationalists is just wrong, one side killed, the other didn’t.

3. The white nationalists chanted things such “blood and soil,” and “Jews will not replace us.” While this is of course horrible, their last chant is the most chilling. “Heil Trump.” (1 ) Of course, no politician is responsible for the acts of all of their supporters. But when a group of Nazis favorably equates the President with Hitler, the President owes it to the nation and himself to denounce the group and make it totally clear that he does not represent that group or its interests.

To his credit; Trump denounced the protestors by name; on a Monday, 48 hours after the murder of Heather Heyer, after just about elected Republican and Democrat in America urged him to.

That denunciation lasted less than a day. Today, Trump held a press conference about something (maybe infrastructure), but during questions, Trump said ” I’ve condemned neo-Nazis. I’ve condemned many different groups. But not all of those people were neo-Nazis, believe me. Not all of those people were white supremacists by any stretch. Those people were also there because they wanted to protest the taking down of a statue, Robert E. Lee… But you also had people that were very fine people on both sides… You had many people in that group other than neo-Nazis and white nationalists. O.K.?”

The aim of the alt-right, is to dress up an old hatred in new clothing, and sell it as respectable. They can’t fool me, and I doubt they’re fooling too many people. But it sure looks like they may have fooled the President. Fooled him into thinking that a lack of tattoos somehow means they’re not Nazis. President Trump is either woefully uninformed about what happened (despite saying 11 times that he waited until all the facts came out to make a statement), or he is endorsing some forms of white nationalism and white supremacy.


I have this illusion, about myself. Its that there is nothing left that Trump can do that can shock me, that he has no further capacity to disappoint me. And every so often, he does. Today was one of those days. Its hard to believe, but I guess I still thought that Trump had some decency left in him, that he could recognize evil in most obvious forms.