Sunday, January 31, 2016

How does this work out on a balance sheet?

In a guest post at Moneyness, Mike Sproul writes:
The bank's IOU will circulate more easily than your IOU, so we commonly talk as if the bank has created money. This is not quite right because the bank is not short in dollars on net. The bank went short in dollars as it issued its IOU, but it took an offsetting long position in dollars when it accepted your IOU. The bank is therefore neutral in dollars, while the borrower is short in dollars. This is why it makes sense to say that borrowers are the original issuers of money, while the banks only help out by putting their name on the money.

I thought Mike's conclusion -- that borrowers are the original issuers of money -- was pretty interesting. But I don't think in terms of balance sheets. So I ask: What say ye?

In follow-up comments, JP Koning asks

Why is the borrower short in dollars on net? A borrower goes short in dollars as it issues its IOU to the bank, but takes an offsetting long position in dollars when it accepts the bank's IOU.

Koning's question: I go to the bank and borrow a million dollars. The bank has my IOU, so I'm short a million bucks. But the bank puts a million in my checking account, so I'm also long a million bucks. Netting it out, I'm just even: not long, not short.

But of course, I borrowed the million because I plan to spend it. And when I spend it I won't have the million anymore. I'll have a really, really nice computer setup instead. So then I'll be short in dollars, and long in hardware. Mike Sproul's answer:

The borrower only holds the bank's IOU momentarily. Once the borrower buys a house, the borrower is short in dollars and long in houses.


What if the borrower buys a dollar-denominated bond instead of a house?

My head starts to hurt. But there's only one more part to the exchange of comments, and Sproul is pretty funny in the end.

Not the funny part:

The origination process is simultaneous in the same way that minting coins is. The "originator" brings an ounce of silver to the mint, and the mint stamps it into a $ coin. Both had a hand in the process, but the originator provided the silver, while the mint only put his name on it.

I'm tempted to say the borrower (or the guy with the bag of silver) is the "issuer" and the bank does the "monetizing" -- which is just about where Mike Sproul started: "borrowers are the original issuers of money, while the banks only help out by putting their name on the money."

Sproul's conclusion is that it's not lending that creates money, but borrowing and lending together. I can live with that.

Saturday, January 30, 2016

Monetary imbalance arises equally for fiat and for gold

David Glasner, commenting on an op-ed piece by Eric Rauchway:
... I stopped nodding my head in agreement with Rauchway when I reached the fifth paragraph of his piece.

Under a gold standard, the amount of gold a nation holds in bank vaults determines how much of its money circulates. If a nation’s gold stock increases through trade, for example, the country issues more currency. Likewise, if its gold stock decreases, it issues less.

Oh dear! Rauchway, like so many others, gets the gold standard all wrong ...

The gold standard operates by fixing the price of currency at a certain value in terms of gold .... The amount of currency under a gold standard is therefore whatever quantity of currency is demanded at the fixed price. That is very different from saying that a gold standard operates by placing a limit on the amount of currency that can be created.

There's more. I'm leaving out stuff about Ted Cruz and stuff about reserve requirements and interest rates in the gold standard era. Good stuff. But the point I want to focus on right now is this: The gold standard did not limit the quantity of currency. That seems to be what David Glasner said.

I was wondering how the quantity of currency could increase beyond the limits set (or not set) by gold. Wondering what it means, that the amount of currency might be "whatever quantity of currency is demanded at the fixed price." I just don't see people lining up to pay an ounce of gold to get $35 of paper. I just don't see it. But then it struck me.

In those days it worked pretty much like it was working for us, while our economy was still working. You'd go to the bank and take out a loan and get some paper money and have money to spend. You didn't pay an ounce of gold for each $35 you got. You just took out a loan.

In the days of the gold standard, the money authority was not able to create base money on demand. These days the money authority can create base money on demand -- "from nothing", as people like to say. That's one difference between then and now.

But that difference is small, I think, compared to this one similarity: In both cases, then and now, it is the money created by borrowing, growing until it gets unsustainably large compared to the monetary base, that is the source of the troubles.

Friday, January 29, 2016

I'm with Irv

From, The Problem Isn't Overproduction — It's Malinvestment:
Austrians do not contend that there cannot be a surplus of some goods. Of course, there can! But we know that a surplus of some goods means that there is a scarcity of others.

From Irving Fisher, #13 in The Debt-Deflation Theory of Great Depressions:
The classical notion that over-production can only be relative as between different products is erroneous. Aside from the abundance or scarcity of particular products relative to each other, production as a whole is relative to human desires and aversions, and can as a whole overshoot or undershoot the equilibrium mark.

Thursday, January 28, 2016

Funny guy, Irving Fisher

From The Debt-Deflation Theory of Great Depressions (1933) by Irving Fisher:
5. The innumerable tendencies making mostly for economic dis-equilibrium may roughly be classified under three groups: (a) growth or trend tendencies, which are steady; (b) haphazard disturbances, which are unsteady; (c) cyclical tendencies, which are unsteady but steadily repeated.

6. There are two sorts of cyclical tendencies. One is "forced" or imposed on the economic mechanism from outside. Such is the yearly rhythm; also the daily rhythm. Both the yearly and the daily rhythm are imposed on us by astronomical forces from outside the economic organization; and there may be others such as from sun spots or transits of Venus. Other examples of "forced" cycles are the monthly and weekly rhythms imposed on us by custom and religion.

The second sort of cyclical tendency is the "free" cycle, not forced from outside, but self-generating, operating analogously to a pendulum or wave motion.

7. It is the "free" type of cycle which is apparently uppermost in the minds of most people when they talk of "the" business cycle. The yearly cycle, though it more nearly approaches a perfect cycle than any other, is seldom thought of as a cycle at all but referred to as "seasonal variation."
5b: haphazard disturbances are today called "shocks".

5c: by "steadily repeated" I think Fisher means happening over and over, not on a regular timetable.

6: it seems to me that these two sorts of cyclical tendencies have been generalized into "exogenous" and "endogenous".

7: this one cracks me up.

Wednesday, January 27, 2016

"identifying the underlying structure and forces"

Science is made possible by the fact that there are structures that are durable and are independent of our knowledge or beliefs about them. There exists a reality beyond our theories and concepts of it. It is this independent reality that our theories in some way deal with. Contrary to positivism, I would argue that the main task of science is not to detect event-regularities between observed facts. Rather, that task must be conceived as identifying the underlying structure and forces that produce the observed events.

Tuesday, January 26, 2016

Like lovers -- inseparable, except sometimes

Graph #1: Federal (blue) and Financial (red) Debt
Well at least they get back together, there at the end.

Monday, January 25, 2016

Government Share of GDP

Yesterday we looked at GDP and components. I said it looked like the government's share of GDP had been falling since 1967. Another look:

Graph #1: Government Share of GDP
Decline from 1967 to 1998. Thirty years.

But "Government Consumption Expenditures and Gross Investment" is only one measure of government spending. Another measure I've run across is "Federal Government: Current Expenditures". Here's a comparison of the two:

Graph #2: Two Measures of Government Spending
They look pretty close until the crisis. Here's the red line as a percent of the blue:

Graph #3: Federal Government: Current Expenditures as a Percent of
Government Consumption Expenditures and Gross Investment
Oh! Not as close as I thought. Especially in the 1950s and '60s.

I need to see another comparison. I want to bring forward Graph #1 and update it by adding the red line, current expenditures as a percent of GDP:

Graph #4: Government Spending as a Percent of GDP, Two Measures
Well, the blue line shows decrease from 1967 to 1998. But the red line shows increase from 1965 to 1982.

Apparently, one can argue either way about government spending.

Sunday, January 24, 2016

With any luck, I'll embarrass somebody today

I was looking for the original paper on the Phillips curve from the November, 1958 issue of Economica. The Wiley Online Library has it here as a 17-page PDF.

From the end of the first section:
The purpose of the present study is to see whether statistical evidence supports the hypothesis that the rate of change of money wage rates in the United Kingdom can be explained by the level of unemployment and the rate of change of unemployment, except in or immediately after those years in which there was a very rapid rise in import prices, and if so to form some quantitative estimate of the relation between unemployment and the rate of change of money wage rates.

The periods 1861-1913, 1913-1948 and 1948-1957 will be considered separately.

The PDF includes these graphs:

Figure 1 by A. W. Phillips

Figure 9 by A. W. Phillips

Figure 11 by A. W. Phillips
Three graphs covering a span of nearly 100 years, from 1861 to 1957.

The same Google search turned up mtnotes_phillips.pdf (13 pages). It appears to be a paper from an Economics 403 course on monetary theory and policy. The parent page says This course requires Eco 212 (Principles of Macroeconomics). It's definitely not a beginner's class.

From part 1A of the pdf:
Phillips, an economist in Britian, plotted inflation vs. unemployment in Britian in the 1960’s.

I don't know what he did in the 1960s. But in 1958, the New Zealand-born Phillips published a study of unemployment and wage inflation covering the period from 1861 to 1957. The 1960s came later.

From part 1B of the pdf:
Primarily the data given by Phillips is just a few periods in the 1960s.

And again:
The original Phillips curve used a very small data set: just a decade.

No. The original Phillips curve looked at the period from 1861 to 1957.

Look, everybody makes mistakes. But the mtnotes pdf -- maybe "mtnotes" is short for "empty notes" -- the mtnotes pdf completely ignores the century of empirical data in which Bill Phillips found the relation that he wrote up in his 1958 paper.

No, the mtnotes guy doesn't just ignore that century of data. He makes up a story to convince us that there was no century of data. He tells a lie.

Is this the right way to do economics? The best lie wins?





Personal Consumption Expenditures, Billions of Dollars, Seasonally Adjusted Annual Rate (PCEC)

Gross Private Domestic Investment, Billions of Dollars, Seasonally Adjusted Annual Rate (GPDI)

Net Exports of Goods and Services, Billions of Dollars, Seasonally Adjusted Annual Rate (NETEXP)

Government Consumption Expenditures and Gross Investment, Billions of Dollars, Seasonally Adjusted Annual Rate (GCE)

I added up the four components of GDP, showed em in blue. Made it a wide line.

On top of it I put GDP, narrow line, red.

Good match:

Graph #1: GDP (red) atop the Sum of Components of GDP (blue)

The four components:

Graph #2: The Four Components of GDP
Blue, consumption, is the biggest part. The government share is just about the same size as investment. The green one, net exports, is negative in recent years.

Rearranged and stacked:

Graph #3: The Four Components of GDP as Stacked Percentages of Total GDP
The lowest line, purple, is the government share of GDP. In the neighborhood of 20% but quite clearly declining since 1967. Since 1967 -- that surprised me.

From the purple up to the blue is personal consumption, the largest share of GDP. From the blue up to the red is investment, gross private domestic investment.

Topping us off at 100% of GDP is net exports. We have a trade deficit, so net exports is a negative value. In order to get the green line up to 100% I put a minus sign in front of it. From the green line down to the red is net exports. Or, from the green line up to the red for the first few years shown, when the trade balance was in our favor.

Saturday, January 23, 2016

Take a system that works, and don't change it much

I don't like change. I don't like getting a new version of Windows. I don't even like buying gasoline at a new place. I mean really, I don't like change.

It's not that I think everything's perfect. I accept change when it brings obvious improvement. But a lot of change is change for the sake of change, ribbons in place of menus in AutoCad and Excel. And a lot is change to make you want to buy the new version. Like your phones. Or like my parents' new car every three years, back in the golden age. It was expected behavior. You were expected to buy a new car every three years. You're expected to buy a new phone every couple years.

Me, I don't have a phone.

Anyway, I guess you'd say I like change to be minimal. For example, before I retired I did AutoCad drawings for a manufacturing firm. Drawings and parts lists. Parts lists by hand. For every set of drawings, there was a list of wooden parts, and a list of metal parts, and a list of plastic parts. And a second list of wooden parts. And the four lists had to be compiled to get summary lists of material requirements that were used for ordering. And it was all done by hand. For every set of drawings.

Mundane repetition drives me crazy.

It was all done by hand when I started. It was all automated when I left. But here's the thing. I took each existing form, one at a time, and made a form in Excel that looked just like it. I made the computerized forms look just like the forms that were done by hand. Minimal change. It wasn't like I thought I had a better idea of the information they needed to gather. I was happy to start with a system that had been in place for years, a system that had been working for years.

Minimal change. Next, I figured out how to automate those forms. I got AutoCad to gather information from the drawings and put it into those Excel forms. And I got Excel to take that information and compile it on the summary forms. I got it all working. And when I was done, we were clicking a button to create a list, instead of writing out material requirement lists by hand.

The key to the whole thing was minimal change. Take a system that works, and don't change it much. Just make things better by eliminating the tedium.

I don't like change.

The economy is a system that had been in place and working for many years. It's not working now, I'll give you that. But I'm not the one looking at the economy as it is now and demanding change. I look at the economy when it was working and ask what made it go bad.

Like you, I want to make things better. But I don't want to throw away everything that seems to be a problem. Minimal change. I want to fix the one thing that created all the other problems. This is where we differ, you and I.

The one thing that created all the other problems is a money problem: the growth of private sector debt. Every time I say that, somebody says I mean private debt should be zero. No. I'm the minimal change guy, remember? Anyway, we couldn't eliminate private debt if we wanted to. But we can change the policy target. We can change the target from "having more debt" to "having less debt".

But we can't even do that until we change the mindset that says "we need more debt for growth".

The mindset has to be this: There is an optimum level for debt in our economy. Too much debt is no good. Too little debt is no good. We need the "just right" level, like Goldilocks. The just-right level of debt is the level that least hinders and most encourages economic growth. This is the mindset we need.

See how I put "hinders" first there, before "encourages"? People tend to forget that debt hinders growth. Policymakers forget. Or maybe they don't know. Maybe they think we get something for nothing when we use debt? Nah, that's just silly. It doesn't work that way. There's no such thing as a free lunch.

So maybe we are at peak oil. Or maybe growth cannot go on forever, and we just happen to be at the place where growth stops, the place where "forever" reaches its untimely end. I don't think so, but I don't know about such things. What I do know is that when there is a problem with money, it looks like a problem with everything money touches -- oil, for example, and toilet paper, and cabbage patch dolls, and exponential growth in general.

I think most of our product-shortage problems are results of problems of the money. I could be wrong about product shortages. But I'm not wrong about the problems of the money. What we need to do is fix the problems of the money, first. And after that we will be able to tell if there really is a peak oil problem, a drug shortage, a food shortage, and all the rest.

Take a system that works, and don't change it much. Find the first thing that went wrong, and fix that. See if the other problems start to go away on their own.

I thought I wrote about this before, but maybe not. At work back in the 1980s, the accountant was showing me something on the TRS-80 Model II computer -- you know, the one with the 8-inch floppy disks? Yeowza. The program was written in BASIC. This was back before GW-BASIC and way before VBA. Long ago. His program crashed.

The guy looked at the code where the error was, and found the variable he wanted to check. Then he typed in three or four commands to check a value and change it or whatever he was doing, I don't really remember now. But I do remember that there was an error in the first command he typed. The computer kicked back an error message. The accountant didn't see it because he was looking at the plan in his head instead of the words on his screen. So he typed the next command, and two more, and he got a result that didn't make sense to him.

It didn't make sense to him after the last command he typed. But it didn't make sense to the computer after the first command he typed, the one where he got an error. It's like that with the economy too. We have to find that first problem. And we have to read the error message.

The accountant missed the problem that returned an error message. He finally realized there was a problem after the last command he issued. But by then, the original error message had scrolled off the screen. It was too late to figure it out, then.

Hell no, I didn't tell him. The evidence already scrolled off the screen.

We have to find the original problem in our economy. The other problems are consequences of the first one. Okay, the economy isn't the same as a computer. Maybe not all of our economic problems are consequences of the excessive accumulation of debt. But most of 'em are. The only way to solve those problems -- and the only way to discover which ones are not consequences, but actually problems in their own right, is to go back to the beginning, back to when things were good, and start looking for the first error we might have missed along the way.

When we finally do that, we will discover that the first error was to allow the excessive growth of private sector debt. That's how we'll know which problem to fix first.

But we better do it soon. The screen is scrolling.

Friday, January 22, 2016

Think bigger

Yesterday I said:

An increase in debt ... creates a "boost" for the economy because it is extra money put into the spending stream. Paying down the loan creates a "drag" for the economy because money is taken out of the spending stream.

It's putting money into the economy that creates a boost. Our method of getting boost is to rely on debt. We borrow and spend.

The trouble with that is, it's only the increase in debt that creates a boost. Once the money's in the economy for a while, it stops having a boost effect. And then nothing's left but the drag created by existing debt.

As time goes by, new debt becomes existing debt. That makes the drag created by existing debt even bigger than it was. It's not like we got something for free. Debt has payback consequences like everything else in life. You know it's true.

The way the economy works -- the standard set by policy -- is that we borrow and spend and this boosts the economy. "We" being not only government, but also consumers and businesses. We borrow and spend and boost the economy. And then, whenever we get a little too much boost and the economy starts to "overheat", they raise interest rates.

Like now: The Federal Reserve just raised interest rates.

I guess the economy was overheating again. I must have missed it.

Anyway, that is the problem. No no, not the overheating. Well yeah, they could be wrong about that too. But that's not the real problem.

The problem, really, is that the different pieces of policy don't work well together. Policymakers encourage borrowing until they start to worry about prices going up. And then they raise the price of borrowing. It happens so often and we've heard about it so many times that it seems to make sense. It doesn't make sense.

To boost the economy, policymakers encourage borrowing. To reverse that process, policymakers have to encourage the repayment of debt. That's all they'd have to do, to prevent inflation: get people paying their debts a little quicker.

We don't have to discourage borrowing. Let borrowing continue. Let economic growth continue. Just put the right kind of limit on it. When you encourage borrowing, the economy ends up with extra debt. To fix that, we need a policy that accelerates the repayment of debt.

Fight inflation by paying down debt. Two birds, one stone.

It's not a difficult concept. It's just that maybe you never thought about it. Like the policymakers.

It's time to think bigger.

Thursday, January 21, 2016

The drag created by the cost of interest is greater than the boost arising from new borrowing

Debt is an asset. And debt is a liability. If I owe you money, for you it's an asset; for me, a liability. But it's important to remember where the problem arises. When debt becomes a problem, it is because the liability became a problem. If the payments take too much of my income, it's a problem.

When debt as liability becomes a problem, debt as asset becomes a problem: If I can't pay you the income you're planning on, it's a problem for you, too.

The simple and obvious solution is to make sure debt never gets big. That way, if debt ever does become a problem, at least it's not a big problem. But maybe that's too simple. Nobody seems to think it's a good idea. Everybody seems to think we should maximize debt right up to the absolute limit and then maybe just a little more, to see if we're there yet.

And then we have a problem.

An increase in debt -- or a new use of credit; it's the same thing -- an increase in debt creates a "boost" for the economy because it is extra money put into the spending stream. Paying down the loan creates a "drag" for the economy because money is taken out of the spending stream.

As a rule, though, the accumulation of debt increases from year to year. There's an increase in debt every year. As a rule. Crisis creates an exception to that rule, but in a normal economy debt increases every year. If I borrowed 80 last year and pay it back this year, but also borrow 100 this year, then there was a net increase of debt. And that creates a boost to the economy.

To see the boost we only look at the increase in debt. Not the total debt.

So I want to look at what we gain by borrowing money. I want to look at the boost and ignore the drag. And -- oh, but the interest. We do have to pay the interest on all that debt we're ignoring. That cuts into our boosted spending.

New borrowing is a boost, and interest payments are a drag, and we'll otherwise just ignore debt for now. I put that on a graph:

Graph #1: Annual Increase in Total Debt (blue) versus Total Interest Paid (red)
The interest on debt is generally larger than the increase in debt. It's a net loss to the transaction economy.

The next graph counts new borrowing as a plus and interest costs as a negative. It shows the difference. When the blue line goes below zero, interest takes more money out of the economy than new borrowing puts in:

Graph #2: Money that New Lending Added to the Economy minus Money Paid as Interest
Most of the time, interest takes more money out of the economy than new borrowing puts in. If that's true, it explains why economic growth has been getting slower since the 1970s: The drag created by the cost of interest is greater than the boost arising from new borrowing.

The obvious exception occurs between 2002 and 2006, where the blue line rose to a sharp peak. New borrowing for that brief time was increasing at such a rapid pace that it managed to produce a net gain for the economy.

And yes, we're still paying for it today.

Wednesday, January 20, 2016

Those who do not learn from history...

You think interest is a good thing, don't ya? You want the nest egg.

Who wouldn't?

Really. I mean, who wouldn't? Everyone wants financial security.

Suppose we run with that. Suppose everybody saves every dollar of income they ever get. What happens then? What happens to GDP?

GDP goes to zero.

What happens to income?

Income goes to zero.


Suppose instead, there is just a little inequality. One guy gets ahead of everybody else. And that one guy saves a little out of every paycheck. What happens?

Eventually, income goes to zero.

You don't believe that, do you. But if only one guy is saving, eventually that guy ends up with all the money. Or, with so much of the money that the economy can no longer function. And then income goes to zero. Or you know, we have a crisis. Call it zero.

Call it unacceptable.


Everybody wants a nest egg. Everybody wants to save. Suppose that happens. What's the result?

There is an increase of savings, relative to everything else.

Suppose this continues for a long time, a hundred years say. What's the result? There is, after a hundred years, a very large amount of money set aside in savings. And not to say people are greedy, but right is right, and nobody wants somebody to come in and steal their savings. Nobody wants that. And nobody wants their savings eroded by inflation. That's not right, either. Yeah, those bristly bums that don't have savings, they might call for inflation. But they're just trying to steal my money. Can't have that. We have to prevent inflation. We have to limit the quantity of money to prevent inflation.

See? We put so much money in savings that it starts to cause problems in the economy, on the way to zero income but still yet a long way off, so much money in savings that the transaction economy begins to suffer because there is not enough money for transactions. And then we say: Oh, no! Don't you dare print more money! You'll erode my savings!!

See how it works?

But it's natural to want a nest egg. And it's natural to want to protect it, once you have it. So this problem will always arise.

Fall of civilization.

Those who do not learn from history...

Tuesday, January 19, 2016

Because I had to look

Jared Bernstein's recent productivity post includes this graph of price changes in IT equipment:

Graph #1
Note the low in the good years of the 1990s. When I see something odd in the 1990s I always think of the debt-per-dollar graph:

Graph #2
That's a new version of my DPD graph. TCMDO is "discontinued" at FRED. This graph uses the two series that together form its replacement. And I used the M1 measure with "sweeps" money counted in. Doesn't go very far back in time, but then that's the tradeoff: This is the "new" version of the graph. Even the oldest data on it is still new.

And then, what the heck, I re-created the DPD as an Excel graph and brought in Bernstein's IT Price graph as a background image. Moved and sized the background image to make the dates line up with my DPD dates. And there you go:

Graph #3
Any relation between the two graphs? Eh...

Maybe. I can imagine that the red and blue run parallel from 1984 to 1994, one smooth, one not, but parallel. Then the big drop in the blue, big price drops for half a decade, these follow after the down-jog in financial cost (red), just as soon as the economy started growing again.

After the 2000-01 peak, the red and blue run parallel again to the crisis. After the crisis, no relation.

I can imagine it.

This graph makes me want to look at other price patterns (other than IT) compared to DPD. I don't expect to find much. But maybe I'll take a look... I should look. My argument is that we had the good years of the 1990s because of the 1990-94 reduction in financial cost that the red line shows. I should expect to find price pattern changes in lots of sectors. Maybe not all as substantial as in the booming IT sector, but visible nonetheless.

Monday, January 18, 2016

It's a big deal, productivity

One thing leads to another.

I found a really good post on productivity slowdown by Jared Bernstein.

"It’s hard to know what drives productivity trends up and down," Bernstein says,

but I’ve got a couple of theories. Investment in productive capital is a known driver of productivity growth, and its slower growth rate in recent years shows up as one reason for productivity’s deceleration.

Every time I read something like that, I want to go to the graphs. Soon enough. First, though, let me finish reading Bernstein's paragraph:

But that just begs the question: why the slowdown in investment? (My other theory is that there’s a full employment productivity multiplier—full employment drives firms facing higher labor costs to find efficiencies they otherwise didn’t need to maintain profits. I won’t get into that here but it suggests what I believe to be an important linkage between productivity growth and persistently weak labor demand.)

"Full employment drives firms facing higher labor costs to find efficiencies", he says. Interesting thought. Reminds me of something Matias Vernengo said:

In Jeon and Vernengo (2008) we suggest that labor productivity is endogenous, explained essentially by the expansion of demand ...

Labor productivity increases with growth, Vernengo says. Bernstein and Vernengo are on the same page here. Vernengo:

... it is the weak recovery ... that explains the poor performance of productivity.

J.W. Mason has more:

Now, whether demand actually does matter in the longer run is hotly debated point in heterodox economics...

Which brings me to this recent article in the Real World Economic Review. I don't recommend the piece -- it is not written in a way to inspire confidence. But it does make an interesting claim, that over the long run there is an inverse relationship between unemployment and labor productivity growth in the US, with average labor productivity growth equal to 8 minus the unemployment rate. This is consistent with the idea that demand conditions influence productivity growth, most obviously because pressures to economize on labor will be greater when labor is scarce.

Seems to me an important idea that is too slowly gathering steam.

Soon enough? Jared Bernstein says

Investment in productive capital is a known driver of productivity growth, and its slower growth rate in recent years shows up as one reason for productivity’s deceleration.

So I went to FRED looking for productive capital and its slower growth rate.

Graph #1: Capital Stock in 2005 Dollars, 1950-2011
The slower growth in recent years that Bernstein mentions, maybe we can see that where the line flattens out somewhat after 2008. Not much to see, as the graph only runs to 2011. I assume the slower increase continues even now, but I don't have numbers to show you.

Suppose we put the Graph #1 data on a log scale. That way the slope of the line more accurately represents the growth of the capital stock:

Graph #2: Capital Stock (in 2005 Dollars) on a Log Scale, 1950-2011
The numbers on the vertical scale are no longer equally spaced. That's because it's a log scale. That's how they make it so that a change in the line represents a change in growth of (in this case) the capital stock.

On Graph #1 the line seems to be generally curving upward until the flattening after 2008. But Graph #2 shows a trend of slower increase for a much longer period. The line runs along a straight path until the mid-1970s. Then it flattens somewhat and runs uphill at a slower pace until 2008, where it flattens more.

So let's get serious and look at the year-on-year growth of capital stock:

Graph #3: Capital Stock (in 2005 Dollars) Year-on-Year Growth
Ignore that big drop from 7½ to 3½ right at the start. Consider the trend of the rest of the data. Looks like: increase to the mid-1960s ... decrease to 1991 ... increase to 2000 ... and decrease to the end. Maybe you can see some of that in Graph #1 or #2  if you look hard enough.

I downloaded the numbers, trimmed off the opening drop, and put a Hodrick-Prescott curve on the graph in red.

Graph #4: Capital Stock Growth Rate, with H-P Trend
Increase to the mid-1960s ... decrease to 1991 ... increase to 2000 ... and decrease to the end.

Hold on now. Hodrick-Prescott is quirky. In an old one at Stephen Williamson's, Angelo remarks:

The HP filter's estimate of the trend is not very reliable at the endpoints of the sample.

Williamson replies:

Yes, exactly. I know some people make an endpoint adjustment to counter this ...

I've run into that problem myself, as have others. But it's not just an endpoint problem. It's also a problem at trend changes. More subtle and harder to find, but no less of a problem.

The H-P calculation smooths out a line and shows the trend of that line. I find it very useful, sometimes. Sometimes not so much. On Graph #4 I trimmed off the opening drop because if I didn't, the red line started high, dropped quickly, and turned slowly. It took till around 1958 (maybe after, I forget) for the red line to get back where it is in the version shown here. That one anomalous data point in 1951 took the uptrend out of the whole 1952-1965 piece of the red line.

More generally, each data point on the red line is figured from several of the nearby data points on the blue line. The red line is like a moving average. (Maybe it *is* a moving average, I'm not sure.) When you're near the start or end of the data series, data points are missing. The data before the start and the data after the end are missing. That throws off the H-P line. That's the endpoint problem.

If there is a distinct trend change somewhere in the middle, between start and end, you have a similar problem. The data from one trend is missing, because there's a different trend after the change. The values from the two trends get "moving averaged" together. The H-P calculation gives you one nicely smoothed line. But if you should be looking at two distinct trends, the H-P doesn't show it.

Everybody that talks about productivity talks about the good years of the 1990s. Those years stand out from the rest because they were so good. They stand out because the trend was different in those years. But you can't see that on Graph #4, because the two trends are merged together by the H-P calculation.

Here's that last graph again, revised to show two separate H-P trend lines. I ended the one trend and started the other in 1992. If there was a long slow moldering of capital growth, followed by a brief revival in the 1990s, it would look something like this:

Graph #5: Better: Capital Stock Growth Rate, with Two H-P Trends
A similar two-trend pattern overlaid on the FRED graph:

Graph #6: Capital Stock Year-on-Year Growth, with Two Linear Trend Lines
Capital stock growth slowed from the mid-1960s to the 1991 recession, then increased during the 1990s. Note that the increase of the 1990s did not become unusual until around 1994, when the increase surpassed the trend of declining peaks of the earlier period.

Just for the heck of it now, I'm adding my debt-per-dollar ratio to that last graph:

Graph #7:  Capital Stock Year-on-Year Growth, and the Debt-per-Dollar Ratio
The debt-per-dollar ratio (green) goes generally upward until the crisis. Except for three years or so in the early 1990s, that is. That is when capital stock growth had not yet broken through the barrier shown by that downsloping red line. And then, just at the moment the uptrend of debt resumed, capital stock growth broke through that barrier.

It was the brief decline in debt-per-dollar that gave impetus to capital stock growth.

"Investment in productive capital is a known driver of productivity growth," Jared Bernstein says. So, "why the slowdown in investment?"

In a handful of graphs I have taken you from a picture of slow capital growth in recent years, to a long-term decline interrupted by recovery in the 1990s. We have looked at this decline in some detail. And I have offered an explanation for the decline and the brief recovery.

Jared Bernstein's "other theory" has to do with labor cost. My theory has to do with the cost of the medium of exchange, financial cost deeply and widely embedded in the economy. For both of us, cost is key.

Sunday, January 17, 2016

An economy that relies on debt for growth

An increase in debt leads to an increase in spending, economic activity and, hopefully, growth as well. But an increase in debt makes the existing accumulation of debt bigger. And the existing accumulation must be serviced.

The cost of maintaining debt is a drag on spending and economic activity and growth. The cost reduces growth arising from an increase in debt.

An economy that relies on debt for growth needs increases in debt that are big enough to offset the whole cost of existing debt -- and bigger even than that, to induce growth.

As time goes by, ever-larger increases in debt are required to induce growth. This is why the productivity of debt declines. It explains the decline you see in this "Debt Saturation" chart from Economic Edge:

Source: Christopher Rupe and Nathan Martin via Joe Weisenthal
It also explains why Federal deficits got so big. And it explains why we have so much private sector debt.

Saturday, January 16, 2016

Keynes on Effective Demand

I looked at this before and got lost in it as I recall. Found a quote in chapter six of The General Theory:

Furthermore, the effective demand is simply the aggregate income (or proceeds) which the entrepreneurs expect to receive, inclusive of the incomes which they will hand on to other factors of production, from the amount of current employment which they decide to give.

First of all, it depends on entrepreneurs' expectations, which drive them to spend the money to produce the product that, when sold, will generate the income.

Second, the expected income includes the portion of that income which the entrepreneurs will have to fork over to other entrepreneurs. Since that portion is included, and as we are not doing calculations here, I'm going to omit the part about portions of income.

Effective demand is the income entrepreneurs expect to receive from the amount of current employment which they decide to give.

Effective demand is expected income, then. It is the particular expected income that leads to employment.


Effective demand is the demand I expect to see, demand as measured by income, income I get from selling my product, expected income from the employment of N workers.

So the word "effective" refers to what, exactly?

Something has an effect, and something is the effect.

The proposed employment of N men has an expected effect on income. This leads the employer to pick a best-case N. Employment follows. Output and income is generated. Done.

So we have an employer thinking about hiring some guys, as the cause of everything that follows.

Or we have employment as the cause of income. Oh -- but that is Say's law.

Okay. Either way, this is not at all what I thought "effective demand" means. And it's not at all what Adam Smith meant by "effectual demand". I'm sure of that.

Looks like I'm lost again.


I did find something in part two of A multi-sectoral version of the Post-Keynesian growth model (PDF, 26 pages) by Ricardo Azevedo Araujo and Joanílio Rodolpho Teixeira:
Unlike the Neoclassical model, the PKGM considers that neither savings nor technological progress is the variable that drives the growth process. The rationale is that investment is determined essentially by the availability of credit in the financial sector as well as the ‘animal spirits’. Once investment is made effective demand determines output which in turns determines savings.

With a comma? Like this:

Once investment is made, effective demand determines output which in turns determines savings.

Here, the employer goes with his gut, giving N workers employment, whereupon demand is generated -- demand, measured by income -- and yadda yadda savings.

It's the decision to employ N workers that has the effect. I guess that's what Keynes was saying. But again, Say's law follows.

Or maybe the comma goes like this:

Once investment is made effective, demand determines output which in turns determines savings.

And then I got nothin'.

Friday, January 15, 2016

An unconventional view of money

Chris Rupe in a Guest Post at Economic Edge:
This is an unconventional view of money...

I have never liked any of the monetary statistics as compiled and named, M0, M1, M2, MZM, etc. I find them too narrowly focused on the liability side of U.S. bank balance sheets. A lot of money is missing in that view. Much of it is obscured by multiplier effects due to the advent of off-balance sheet entities. Some is not counted. Cash held by foreign central banks for example.

The way I see it, all loans must be originated by U.S. banks whether or not they are held on or off balance sheet. This is a distinctly asset side of bank balance sheet view of money. And ALL of this money is accounted for in the Federal Reserve Z.1 as Total Credit Market Debt. So, I have sometimes referred to this statistic as ‘Mtotal’.

I like it. Not sure I get his reasoning. But Total Credit Market Debt gives a very good indication of how far we have stretched our money. Myself, I like to compare it to M1 money, because M1 is the money we can use to pay back debt. That's an important yardstick, I think. I'm glad to see somebody else thinking along related lines.

The post at Economic Edge is an old one. FRED's "Total Credit Market Debt" series TCMDO has since had a name change -- not "debt" now, but "liabilities" -- and, more recently, has been discontinued altogether. I looked at the replacement. See the last graph here.

Thursday, January 14, 2016

Your grandfather started it

Sometimes I have a thought pop into my head: a one-liner that suddenly makes sense to me. It seems to be a direct quote. But it must be a distillation of something.

Take Keynes, for example. He doesn't often write one-liners. But then I wake up one morning with "Oh, Keynes said you can only either spend money or save it." And in my mind, the you can only either spend money or save it part is what Keynes said -- his words, not mine. And that's how I think of it.

And I've had a couple times when I went looking thru The General Theory for that particular phrase. So I could document it in my writing. Looking till my head hurts, but not finding.

Doesn't mean that the exact phrase is not in the book. But by the time my head hurts and I give up the search, I'm convinced that the exact phrase is not in the book. At that point, I have to think that those particular words are not what Keynes said, but just my memory of it. I don't know if that's correct, but it's what I have to think in order to proceed.

After that, I might say something like Keynes said you can only either spend money or save it. No quotes, so it looks like I'm paraphrasing him. But attributed to Keynes, so I'm not stealing the words, just in case. It's a compromise that lets me move on to the next sentence.


What I did find in The General Theory was this:

Saving, in fact, is a mere residual. The decisions to consume and the decisions to invest between them determine incomes.

and this:

Clearness of mind on this matter is best reached, perhaps, by thinking in terms of decisions to consume (or to refrain from consuming) rather than of decisions to save.

That's gotta be where it came from.

It follows from what Keynes said, that if you want to save, you have to not spend. In other words, your only choices in this matter are to spend or not.

You can only either spend money or save it.


I looked up think of "saving" as "not spending". It seemed to trip google up. I got Don't confuse 'saving' with 'not spending' at USA Today. The link is for documentation only. They've got a pop-up ad that makes me go all Tourette's.

Doesn't take much. Anyway, they say

Saving money and not spending money are not the same.

That obviously contradicts Keynes. How come? They continue:

Say, for instance, you decide to take your lunch to work for the next couple of days to save money. Yet, two weeks later, your savings account is no higher for your effort.

Oh. It's only "saving" if you put it in a savings account. That's what this guy is saying.

The same search turned up Avoiding Spending Is Not The Same As Saving Money by Trent Hamm at The Simple Dollar:

Eliminating a routine that involves spending money needlessly, or substituting a different routine that involves spending less money, is a great way to cut your spending...

But it’s just a single step. Alone, reducing spending cannot bring about positive financial change in your life...

You have to do something financially productive with that money.

Same thing. If I quit buying coffee on the way to work in the morning, it's a noticeable cutback in my spending. But if I spend the money on beer on the way home, I didn't "save" it. I get Trent Hamm's point. And the other guy.

How does this differ from what Keynes said?

Trent Hamm and the other guy are talking to the individual, the reader. They're trying to help people cope with a troubled economy. If you want to save -- if you want to increase your savings -- you have to put money into savings. That's their plan.

Okay. But Keynes is in big-picture mode. He's talking about how an economy becomes a troubled economy.

He's not just looking at our morning; he's looking at our whole day. If I don't buy coffee in the morning, that is a decision not to spend; if I buy beer on the way home, that is a decision to spend. All told, I spent the money. So I spent it on beer instead of coffee. I still spent it.

Talking about spend or not spend, Keynes means spend or not spend. If you spent your coffee money on beer, you still spent it. But if you take that coffee money and choose not to spend it, that's different. It's different, no matter where you put the money, as long as you don't spend it.

And if you look at all of us, not each of us, over fifty years or more, the amount of money the savers among us have chosen to "not spend" is a big number.


Then I Googled keynes "not spending". Another Simple Dollar post came up:

The Paradox of Thrift
by Trent Hamm.

Subtitle: Is Saving Money Bad for the Economy?

Two years ago (in those economic halcyon days before the so-called “Great Recession”), I wrote a short article entitled Is Not Spending Money Bad for the Economy? In it, I largely concluded (by my own logic) that not spending money – in other words, saving it – isn’t necessarily bad for the economy at at all.

It's kinda cute. Now he's equating "not spending" with "saving". And these are the only two posts of his I ever read.

Anyway, Trent Hamm says

John Maynard Keynes ... believed that if everyone saved more money during times of recession, then demand for goods will fall.

And then Trent says

First, when demand falls, prices fall. When prices fall, people are more likely to spend money.

Yeah, in a normal economy, yeah. But when things go haywire, and so many people are saving more money that it causes demand and prices to fall, then maybe it's not a safe bet that the reduced prices will reinvigorate spending. Again, Keynes is thinking big-picture; Trent Hamm is not.


Trent Hamm again:

The second factor – and this is the big one – that makes the “paradox of thrift” fail is that putting money in savings accounts does not remove it from the economy. When you put money in a savings account, it becomes money that the bank can then lend out to businesses. Thus, when more people save, the banks have more resources to pump out to businesses, and when the businesses have more resources, they employ more people, innovate new products, and find new ways to sell.

I crossed out the happy ending. Everybody's economic theory has a happy ending. It's all predictions and bullshit. Don't give me the happy ending. Just give me the argument and let me evaluate your thinking.

The last thing I didn't cross out -- "the banks have more resources to pump out to businesses" -- sounds to me like "loanable funds" theory. I thought like that when I started this blog but I've had it pretty well beat out of me by Greg and others.

I don't really know what "loanable funds" theory is. But the idea that if you put a dollar in the bank, I might borrow it seems to make sense. Maybe that's one way it works (though maybe not at "banks") and another way it works is "lending creates deposits". And clearly, if the bank can create a dollar by lending it to me, then me and the bank don't need your dollar before I can borrow.

I don't know quite how it all works. I'm told banks don't need loanable funds because they do creatable funds. Still, banks are always trying to get depositors. And banks are always trying to get deposits. They must need those deposits for something.

No matter. Let's ignore that part of the Trent Hamm excerpt, too. What we're left with, then, is something like this:

The second factor – and this is the big one – that makes the “paradox of thrift” fail is that putting money in savings accounts does not remove it from the economy. When you put money in a savings account, it facilitates bank lending.

Everybody okay with that?

I'm not done yet. Trent says

putting money in savings accounts does not remove it from the economy.

I disagree. It does remove that money from the economy. That money is removed from the spending stream. It is no longer part of what FRED calls "funds that are readily accessible for spending." (See the "Notes" tab just below their M1SL graph.)

The economy is transaction; money in savings is not in the spending stream; therefore money in savings is out of the economy.

Trent is right of course: Money does come into the economy when the bank lends it to somebody. But it comes in at interest. Thereafter, financial costs in the economy are higher than before.

Now, that may not seem like much. But if you've got 150 million people in the labor force, all of them at one time or another thinking about borrowing money -- and if you've got economic policies that encourage borrowing but do not encourage repayment of debt -- then if you just let the economy do its thing, the little extra financial cost that arises with that first innocent loan eventually becomes a financial burden too great for the economy to bear. And then you have a crisis.

You have to go big-picture here. Don't think about your one loan and the cost of one loan. Think about the loan your grandfather took when he was young. Think about how the cost of that loan added something to his cost of living. If he was in business, think about how the cost of that loan ate into his profits, and how he raised his prices to compensate. And then, think about how when his loan was paid off some years later, he already had two other loans bearing similar financial consequences. And by the time his loan was paid off, maybe 40 million people were deeper in debt than they were when he took out his loan. Big picture.

And then because all those people had increasing financial cost burdens, financial sector income was growing at a healthy clip. And people began to notice that it was better to put their money into financial investment than into productive investment. So finance grew, and grew faster than the real economy grew. Finance grew faster than output. And financial costs grew faster than output, too. Faster than income.

And this didn't stop, because as financial costs grew the standard of living took a hit, and so did productive business. And as the non-financial sector gradually failed, the financial sector became more and more appealing. And finance grew. And output suffered. And productivity suffered. And living standards suffered.

And then one day, the productive sector could no longer support the financial sector. That's when things got bad for finance. And that is called "crisis".

Wednesday, January 13, 2016

Generational Warfare Propaganda from the St. Louis Fed

In a FRED Blog post titled Changing demographics, the Fed's Maximiliano Dvorkin wants us to see this graph:

Graph #1: From the FRED Blog
I see the lines crossing. I see pretty significant swings, the one line up, the other down.

But then I see that the blue line shows the working-age population as a percent of total population. And the red line shows the retirement-age population as a percent of total population. I see that both lines show parts of the population as a percent of the total, and I wonder why they are not both shown on the same axis.

The one line is always 63.5% or more, while the other is always less than 14.5%. These lines are separated by half the U.S. population or more. That fact is hidden from view on Dvorkin's graph.

I put both lines on the same axis:

Graph #2: The Same as Graph #1 Except Both Lines Use the Same Vertical Scale
The demographic problem doesn't look quite so dire now, does it.

A few weeks back I found Circuit's old post on demographics at Fictional Reserve Barking. Circuit says "seniors are, and will remain, a relatively small share of the total population." You can see this in the second graph, but not on the FRED Blog graph.

Circuit says working people support everyone -- themselves, and people too old to work, and people too young to work, and people of working age who are not working. Everyone. He's right.

And as Jim points out, the baby boomers were all "carried" by the rest of the economy once before: from World War Two to the mid-1960s. Those years, as it happens, were the “Golden Age of Capitalism”.

There is an economic problem, but baby boomers ain't it.

If you think Dvorkin's graph exaggerates the demographic problem to the point of dishonesty, you might want to tell him yourself:

Saturday, January 9, 2016

Looking for significance

Start with my re-creation of Sumner's graph:
Step #1: Currency Relative to GDP (blue) and the Interest Rate, à la Sumner

Invert the blue line, as Sumner does:
Step #2: The Blue Line now shows GDP Relative to Currency

Show the blue line as percent change from year ago:
Step #3: The Blue Line expressed as "Percent Change from Year Ago"

And show the red line as change from year ago, percent:
Step #4: The Red Line expressed as "Change from Year Ago, Percent"
The lines now run pretty close and somewhat similar. Close enough and similar enough that I went back and did the above steps a second time, and captured the images for this blog post.

Consider that last graph. That scribble in the middle, the up-and-down red lines just after 1980, that's Paul Volcker fighting inflation. After the scribble, the red and blue lines show roughly equal size* changes, and patterns that I think are impressively similar. Before the scribble there is some size discrepancy, but the patterns still match up pretty well.

I notice that both the red and blue lines tend to have low points when there are recessions. That accounts for a lot of what looks like similarity. So maybe the pattern similarity is not as significant as I first thought.

Why does the blue line start out so much higher than the red? (In the 1950s, I mean.) Blue shows GDP relative to the currency component of money. In the early years after World War Two there were some very high peaks in GDP growth, real growth. So that pushes the blue line up at the start.

To the left of the Volcker scribble, there is a lot of size difference between the red and blue lines. I did say that already. But despite the size difference, it appears that every time the blue goes up the red goes up. And every time the blue goes down, the red goes down. I know it is recession-related, but it is still impressive.

To the right of the scribble also. One big difference between the lines is from the 2009 recession to the end of the graph, where the blue line  drops way low, and then stays well below zero where the red line sits.

The other big difference in that half of the graph occurs in the early mid-1990s where the red line rises to peak, and the blue lags behind. A pretty good gap opens between them. This difference and this gap occur just as the debt-per-dollar ratio regains upward momentum and we enjoy the good years of the 1990s.

Probably not a coincidence.


The behavior of the blue line is more the result of GDP than currency, as I recently pointed out. And the inverse relation between the red and blue lines on Sumner's graph is due more to GDP than to currency. I'm definitely not saying Sumner is right. But the graph is interesting.

* NOTE: The word "size" needs a footnote. These graphs have two vertical axes, one for the red line and one for the blue. This is done so that the graphing program will re-size and re-position the lines, to make comparison easier. It is still valid to say, for example, that the lines are roughly equal in size, or that there is a size difference. But the word "size" does need a footnote.

Thursday, January 7, 2016

Still troubled by Sumner's graph

Following-up on yesterday's evaluation of Scott Sumner's graph.

Here's my re-creation of the graph Scott Sumner showed on 3 January 2016, which I looked at on the 6th:

Graph #1: Currency Relative to GDP (blue) and the Interest Rate, à la Sumner
Here's what Sumner says immediately after showing his graph:
Notice that the two variables tend to move inversely. After 2008, the yield on T-bills fell close to zero, and the demand for cash soared from just over 5% of GDP to just over 7%.

Just to be clear about what Sumner is saying, I marked up a copy of the above graph to show the "just over 5%" and the "just over 7%" and exactly which part of the blue line he is talking about.

Graph #2: A Markup of Graph #1
The bright red oval identifies the region Sumner described. The bright red arrows indicate the top and bottom of that region, and also which axis shows the values. The black arrow shows where the pukey-red line (the yield on T-bills, which I am ignoring today) falls close to zero, where it remains.

Sumner's argument is that "the demand for cash is negatively related to the market interest rate". That is what his graph shows. The demand for cash (blue) falls for the whole left half of the graph, while the pukey-red market interest rate is rising. And for the whole right half, the blue line rises while the pukey line falls. As Sumner put it, "the two variables tend to move inversely."

Yes, what we see on Sumner's graph is the same as he has described. I had some complaints yesterday about his graph. Have no fear, I have those same complaints today. But to be fair and balanced here, I wanted to be clear on what Sumner said about his graph.

That's out of the way now.

Scott Sumner is telling us that interest rates influence how we hold our money. When rates are low, he says, we hold more of our money as cash. And when rates are high he says, we hold less of it as cash. As I noted yesterday, it's a reasonable argument and it makes sense.

But I do graphs. It's what I do. I can't tell you why his graph bothers me. But it bothers me, and I do graphs. So I had to look at Sumner's graph and see if it made sense or not.


That was yesterday's post.

Today ... today I was looking at his graph again because it shows money relative to GDP ... and I also show money relative to GDP, but different money ... and Milton Friedman also showed money relative to GDP -- but a different GDP.

(Look, in Money Mischief, Friedman used "national income" for GDP. But that's not the different that I'm talking about. National income is either the same as GDP or a bit less, it seems, depending who you ask on the internet. But they run together, GDP and national income. There's no big difference between them. Friedman, though, didn't just use national income: He used "real national income", which is national income with the numbers changed to take inflation out. It's what national income would have been, if prices never went up at all. And that's a big deal. That's what I mean by a different GDP. But that's off-topic. Shame on me.)

Today I was looking at his graph again, Sumner's graph, and it was bothering me, again, and it wouldn't leave my mind alone. Here: If Scott Sumner wants to show that we hold more of our money as cash when interest rates go down, he has to show us the money we hold as cash, compared to all the money we hold. That's what more of our money as cash means: our cash as a share of our money.

He doesn't. He shows the money we hold as cash, compared to the size of the economy. Should be, compared to all the money we hold. Like this:

Graph #3: Currency as a Percent of M1 Money (two measures).
The red line is the better measure, in my view.
If you look at a graph of the currency component of M1, relative to M1 money, the currency component is going down since the crisis. Not up, as Sumner says. It goes down.

Currency relative to the money measure provides a good picture of our inclination to keep our money as cash. Currency relative to GDP, the measure Sumner uses, does not.

I have to assume that Sumner used currency relative to GDP on his graph because it gives results that by dumb luck appear to support what he's saying. And that's very nice for him, because apparently most people don't question graphs. Most people take graphs as evidence. I know I usually do.

Maybe that's why I'm so troubled by Sumner's graph, this time.