The wife is away for the weekend...

...and I am bored silly.

Details

I had to interrupt myself. I was writing a post that looks at an old one by Ron Robins. You'll see tomorrow. Meanwhile, this:

Ron Robins' old post is about the growth of debt. Total debt. One of my favorite topics. He tosses out some numbers on debt growth. Then he adds this:

What must be noted is that for the thirty years prior to the late 1970s the credit-to-GDP ratio held steady around 1:1.4.

In other words, about $1.40 of debt, for every dollar's worth of GDP. Pretty low number. Pretty good contrast to Ron's other numbers. But, steady?

You can see evidence of Ron's claim, more or less, in the first graph at ContraHour's old Seeking Alpha post. The graph is a little faint and fuzzy, but it shows total debt at around 150% of GDP in the late 1940s, dropping to around 130% in the 1950s, leveling off around 145% in the 1960s, and rising only slightly in the 1970s -- only slightly, in comparison to debt growth since that time.

ContraHour's Graph

Below is my graph of the same thing. On my graph, the blue line doesn't match the red and gold lines because of changes in the way debt was counted, or something. But it'll give you the idea.

Here's my spreadsheet with sources & numbers.

I added a black box around "the thirty years prior to the late 1970s" to make it easier to focus on the years Ron Robins is focused on in the quote above. So, here's my graph:

Yeah, I guess you could say total debt "held steady" around 140% of GDP for those years. There does seem to be a little bit of "down" and quite a lot of "up" in those trend-lines. But the "up" was gradual, and total debt was relatively low. So let's go with Ron's description: total debt held steady at around 140% of GDP for that thirty-year period.

It is interesting to note that the "golden age" of post-war capitalism lies entirely within the black box. In other words, the years of relatively low debt were very good years for the economy.

Below is another graph with a box around that thirty-year period. This graph is based on the same numbers as the graph above. But this time I didn't just lump all the debt together. I looked at how much government debt there was (relative to private debt) and how much private debt there was (relative to government debt).

Turns out that during the whole period of low and "stable" debt, government debt was decreasing and private debt was increasing. Makes sense, because those were years of good growth, and we use credit for growth, so that you would expect private debt to be on the increase.

What happened, though, is that private debt just kept increasing until the cost of that debt became too much for the economy to take. And that killed off the golden age.

Either that, or total government debt just got too low and somehow that's what killed off the golden age.

I dunno which. But that's what the graph shows.


To Ron Robins I have to say: I like your focus on total debt. And yeah, Ron, I guess you can say total debt (relative to GDP) was low and stable during that thirty-year period. But actually the government's share of that debt was falling the whole time, and the private share was increasing.

Important details.

Maybe it's different in Canada

At Worthwhile Canadian, Frances Woolley considers the Harper government's Task Force on Financial Literacy. "I regard the the whole financial literacy exercise with some cynicism," she says. "Moreover..."

Moreover, financial literacy campaigns frame excessive debt or insufficient savings as a financial problem - expenditures exceed income. Yet, as the the familiar circular flow diagram shows, financial flows are simply a reflection of real flows. (This flow diagram is borrowed from wikipedia).

Framed in real terms, excessive debt and insufficient savings becomes either a problem of "too much stuff" or inadequate incomes: the value of goods and services coming into the household exceeds the value of that household's labour and capital.

Financial flows are simply a reflection of real flows, Woolley says.

When you 'assume' ...


I choke on Wikipedia's flow diagram. Maybe it's just bad choice of words, I don't know. But that label wrapping counter-clockwise around the bottom there -- Factors for production -- is too close to "Factors OF production" not to be mistaken for it. And "factors of production" is not a concept to be used to mean "inputs."

The factors of production are cost categories; the concept gives us a way to look at costs. What's the difference? "Inputs" are used to produce output. "Costs" hinder.

But set the factors aside. I have a bigger problem with the flow diagram.

Wikipedia's flow diagram shows the financial flows in red. I tweaked it to show the financial flows in green. No big deal, right?

Red is credit-use. Green is cash.

My version is self-contained, complete, and sustainable. The original version is not. Oh, it shows the flows to and from the productive sector. But it fails to show the flows to and from the financial sector. And it fails to show the growth of finance.

"Financial flows are simply a reflection of real flows." Harrumph! There are costs associated with finance. So financial flows are a reflection of real flows plus finance charges. And guess what happens when economists ignore the costs of finance.

Oh that's right. We don't have to guess. We're living through it.

Dirk

From Gang8:

Re: [gang8] Back to interest rates

Chris,

I agree that savings and borrowings are an identity at all points indeed but frankly that does not address the issue. That equality is the case for ALL demand and supply in markets that clear (as all markets do by definition in the perfect-competition model), so that argument won't convince believers in that model (aka economists). What you need to show is that saving/borrowing do not vary inversely with interest rates.

In a 2006 paper I researched this issue and found that NO published academic article convincingly shows that higher (lower) interest should decrease( decrease) lending/borrowing. It's an assumption. I wrote:

And indeed “the intuitive feeling with regard to the importance of the rate of interest for credit … is seldom found in empirical econometrics.” (Fase, 1993:99). Lown and Morgan (2006) show that loan volumes dominate loan rates in explaining output. Geanakoplos (2009:9) recently called in Nature for an end to “the obsession with interest rates” and asserts that “regulating leverage, not interest, is the solution for a troubled economy”. Malcolm Knight, General Manager of the Bank BIS noted in a 2006 speech: “The prevailing mainstream theoretical paradigms, enshrined in current textbooks and research, find it difficult to accommodate a significant role for quantitative aggregates over and above that played by interest rates”.
The widespread practice of reducing the study of credit markets to interest rate or credit spreads relies on the perfect-market assumption that prices reflect all information - completely unwarranted in a market which, as Richard has argued, are quantity-rationed by the supply side (banks) due to information asymmetries.

Dirk

"The widespread practice of reducing the study of credit markets to interest rate or credit spreads" is the reason nobody saw the crisis coming. Nobody was looking at the big picture. Nobody was looking at accumulating debt.

//

"What you need to show is that saving/borrowing do not vary inversely with interest rates."

Interest rates go up and down in a cycle that essentially duplicates the business cycle. The accumulation of "saving/borrowing" goes up and down in a cycle that essentially duplicates the long wave.

There are many business cycles in a long wave. There are many ups and downs of interest rates in one long upswing of debt accumulation. Saving and borrowing do not vary inversely with interest rates.

Or maybe Dirk is talking about new additions to saving and borrowing, as opposed to total accumulations. If so, he is concentrating on a very small part of the problem.

//

“The prevailing mainstream theoretical paradigms, enshrined in current textbooks and research, find it difficult to accommodate a significant role for quantitative aggregates over and above that played by interest rates”.

Economists say they need a new "paradigm" but evidently they don't mean it. Dirk and the BIS guy he quoted have pegged the problem.

The thing of it is, when you put finance people in charge of policy, they just can't see debt as a problem.

//

"...completely unwarranted in a market which, as Richard has argued, are quantity-rationed by the supply side (banks) due to information asymmetries."

Quantity-rationed? Dirk makes it sound as if there is a shortage of available credit. Maybe so, since the crisis. Certainly, before the crisis there was not.

I think Dirk is looking at new issue of credit, as opposed to the total accumulation of credit in use, or total debt. It's the same problem he and the BIS guy observed. The accumulation is the problem. It is a cost burden. And it "crowds out" new borrowing. Only when we pay off a dollar of debt does that dollar becomes available to lend again.

When there is economic slowdown, very likely there is a decline in new lending. But there is no automatic decline in outstanding debt. That's how we get into trouble. And then, with the slowdown, outstanding debt becomes even more unaffordable. That's how we get into crisis.

Dirk is concerned about credit-rationing by the supply side. My concern is completely different. I want to see less DEMAND for credit. I want to establish policies that reduce the demand for credit but provide interest-free money to make up the difference.

I want less TOTAL demand for credit, but not less NEW demand for credit. New uses of credit help the economy grow. Old accumulations hinder growth. Both new credit-use and old outstanding debt are part of the total demand for credit. If the total demand for credit is too great, then we must reduce new credit-use, or we must reduce old outstanding debt, or both.

Existing anti-inflation policy reduces the demand for new credit-use, thereby creating an impediment to growth.

Arthurian anti-inflation policy reduces the demand for old outstanding debt, thereby removing an impediment to growth.

Private Issues post-post

In Defense of Theory

I'm just a tad disappointed in my "Private Issues" series. I wanted to look at some numbers and claim there was evidence supporting my view. But it's like there are no numbers to look at.

The FRED GDP observations go back to 1947, mostly; some to 1929. Nothing in the 1800s. Nor does ALFRED go back in time before 1929. FRASER doesn't list GDP at all, that I could see.

The Bicentennial Edition of Historical Statistics lists GNP back to 1869, but even that is after the Free Banking era. The older HS lists national wealth back to 1774, but GNP only to 1929 and National Income back to 1799 but only for every tenth year.

NBER business cycles start with 1854, the last decade of the Free Banking era.

I ended up using Measuringworth's numbers again. I compared "real" GDP of the Free Banking era (1836-1863) to the post-WWII Golden Age (1947-1973, but I used 1946-1973 for the comparison).

For each period I figured the CFPY (Change From Previous Year) of Real GDP. I took the absolute values of those numbers, to make them all positive. And I compared the minimum and maximum changes for the two periods.

Why? Because it seemed a reasonable approach, that's all.

The smallest year-to-year change during the Free Banking era was 44% larger than the smallest comparable change during the Golden Age. The biggest year-to-year change during the Free Banking era was 42% larger than the biggest change of the Golden Age.

I want to conclude that downturns were more severe in the Free Banking era. But the comparison shows only that our economy was more generally active back then. I need another comparison. I would have to look at the downturns specifically.

The numbers show four negatives during the Golden Age -- four years of decline in RGDP during the period. Oddly, the numbers show no negatives during the Free Banking era: No recessions at all. Not in the Measuringworth numbers, anyway.

//

Wikipedia provides a table listing seven U.S. recessions between 1836 and 1863. There were many. And judging by the declines in business activity, they were severe.

Economictheories.org provides Business Cycles History, which refers to

Professor Wesley C. Mitchell, whose work on business cycles is the best known and the most substantial of all the modern works in this field...

And Professor Mitchell himself provides a chart showing "Approximate Duration of Business Cycles" -- Chart 23 from Business Cycles, The Problem and its Setting by Wesley C. Mitchell, one of the "National Bureau of Economic Research Publications in Reprint." (Image at right.)

Wesley Mitchell's chart shows eight cycles within the 1836-1863 period, one more even than Wikipedia. Table 32 in Mitchell's book identifies four of those eight as involving panic. Inflation was well-controlled in the Free Banking era. Panics were not.

//

The Library of Economics and Liberty offers Business Cycles by Christina D. Romer, wherein one may read:

The prewar versions of these series were constructed using methods and data sources that tended to exaggerate cyclical swings. As a result, these conventional indicators yield misleading estimates of the degree to which business cycles have moderated over time.

In other words, according to Romer, the business cycles of the 19th century were not as severe as we have been led to believe.

On the other hand, in Changes in Business Cycles: Evidence and Explanations (PDF, 1999) Romer writes:

The bottom line of this analysis is that economic fluctuations have changed somewhat over time...

...the advent of effective aggregate demand management after World War II explains why cycles have become less frequent and less likely to mushroom. At the same time, however, there have been a series of episodes in the postwar era when monetary policy has sought to create a moderately sized recession to reduce inflation. It is this rise of the policy-induced recession that explains why the economy has remained volatile in the postwar era.

Romer's point is that business cycles are less frequent and less severe after World War II, as compared with before World War I. (But not as much less as people thought.)

So Romer confirms a bit of my Private Issues hypothesis -- at least, she confirms the environment I need for my hypothesis to be valid: the environment where recessions are more frequent and more severe in the 19th century than in the 20th.

But Romer does more than this. She points out that much of the 20th century volatility arises from inflation-fighting by the Federal Reserve. Yes. In the Free Banking era there was no Federal Reserve. And there was no inflation. But there were panics and frequent, severe recessions.

In the 20th century, we had to create recessions on purpose, in order to do the thing that was done automatically (and more severely, and more often) in the 19th.

And to the extent that Fed-induced recessions were less frequent and less severe than Free Banking recessions, the difference has been made up by the greater inflation of the post-war era. Romer confirms my hypothesis.

Romer calls this "effective aggregate demand management." What was it I said?

They pat themselves on the back and attribute the difference to the wisdom of their policies. I say the difference arises from the fact that we are no longer on gold.

But remember: The problem is not whether we are on gold or not. The problem is the imbalance between base money and private-issue money. This problem resolves itself differently when we are on gold and when we are off. But neither getting on gold nor getting off gold is a solution to the problem of monetary imbalance.

Private Issues (5): Conclusion

In the Free Banking era -- which might more aptly be called the Free-For-All Issue of Money era -- inflation was well-controlled, but panics were not.

In the Golden Age that followed World War II -- which was far too short to be called an "age" -- panics were well-controlled, but inflation was not.

The difference is that the one period had a gold standard and the other did not.

The similarity is that in both periods, the disparity between base money and private-issue money existed. Monetary imbalance existed, in both periods.

I propose to solve the problem of monetary imbalance.

Private Issues (4): Growth

Measuringworth provides numbers on Real GDP:

My graph uses a log scale, which accounts for the ridiculous numbers visible on the vertical axis. Anyway, the graph shows slowdowns that begin around 1839 and 1848 and 1857. The spreadsheet shows related peak growth rates occuring in 1835, 1846, and 1855. Clearly these dates are associated with the Panics of 1837, 1847, and 1857.

Looking at a later period of U.S. history, FTAlphaville provides Deflationary periods in US history, with a chart from the St. Louis Fed and these words (also from the Fed):

The vertical lines on the chart mark the approximate dates of the 1890, 1893, 1907, and early-1930s financial crises. These crises led to bank failures and a reduced money supply as depositors withdrew money, preferring to hold cash.

and these:

Inflation, output, and (as noted) the money supply fell during or shortly after each crisis. The international experience has been similar...

Crises are associated with declines in the quantity of money.

I can imagine people reacting -- perhaps to a perceived shortage of gold, created by the relative increase in use of paper money -- by demanding gold, and holding gold, creating the situation where there was in fact a shortage of circulating gold and a shortage of gold backing for paper money.

I can imagine too that people would remember their experience with past business cycles, for several 10-year cycles fit will in a lifetime. In our day, or at least in the post-WWII Keynesian era, business cycles have not been accompanied by crisis¹. In the free banking era, they were. People would start to notice less gold and more paper in their pockets, and their reaction would bring crisis to a head.

NOTE 1: Kindleberger's Appendix B lists no U.S. crises between the Great Depression and 1973. David Moss's Figure 1 (PDF) shows a comparable absence of bank failures.

It is not so easy today to notice a shortage of base money and an excess of credit use, because none of that is gold. But very little of our spending today involves the use of greenbacks, for example, and everyone has a lot of debt.

The crisis is a re-balancing. The economy uses bad debt (and the resulting reduction of total debt) as a way to reduce the quantity of private-issue money. The economy seeks balance between private-issue money and base money. Excess private money is destroyed.

Thus some people say preventing economic collapse prolongs the problem.

During the Free Banking era, crisis reduced the quantity of money and caused deflation. This counteracted inflationary pressures of the boom, and kept prices roughly stable during the period. There may also have been confidence created by the gold standard, just as today there is a lack of confidence generated by the lack of that standard in a time of troubles.

That confidence contributed to price stability whereas our lack of confidence contributes to inflation. However, the problem is not that a gold standard is now absent. The problem is the growing disparity between gold-backed money and unbacked demand deposits, or between gold-backed money and private-issue money, or between base money and private-issue money, or between money and debt.

It is this disparity that creates the condition that is resolved by inflation or by panic.

Private Issues (3): Prices and Panics

In part one of this series, karthikraoa observed that the year-over-year growth of M3 money is very high, around 20%. In reply, David Andolfatto pointed out that "Most of M3 is private money creation." He said "the stock of base money is tiny relative to M3."

I pointed out that this bigger quantity of money "must have much more to do with inflation than base money does," and concluded "that the blame for inflation lies more with private money than with base." David rejected my conclusion, suggesting that "the US 'free' banking era 1836-63" was an era that showed my conclusion wrong.

In part two, I referenced Warren Mosler's observation that "The gold standard was established in the U. S. in 1834." Thus we were indeed on a gold standard during the free banking era, as I imagined in part one. The link to gold during that time prevented monetary imbalances from working themselves out through inflation. Some other route had to be found to relieve the economy's internal pressures.


Based on annual values from Measuringworth, a plot of consumer prices during the free banking era:

The spike at the right is associated with the U.S. Civil War. Apart from that spike, the entire period looks to average out around the CPI level 8, or 8% of the price level of 1982-84. Not really any inflation during this period.

At the left, the CPI peaks in 1837 and falls for half a dozen years. Then it rises a bit and peaks in 1847. Then it is flat until 1853, rises a bit, and peaks again in 1857. A peak every ten years.

In Appendix B of Charles Kindleberger's Manias, Panics, and Crashes -- "A Stylized Outline of Financial Crises" -- several crises are listed for the United States, including speculative peaks in September 1837 and August 1857.

Wikipedia's Economic History of the United States notes that "President Andrew Jackson ... opposed paper money and demanded the government be paid in gold and silver coins. The Panic of 1837 stopped business growth for three years."

At the time, base money was gold and silver coins.

Wikipedia's Panic of 1857 reports that "The Panic of 1857 was a financial panic in the United States caused by the declining international economy and overexpansion of the domestic economy. Beginning in September 1857, the financial downturn did not last long; however, a proper recovery was not seen until the American Civil War."

"Overexpansion" of private-issue money.

There was also a Panic of 1847, in England. This article quotes Jesus Huerta de Soto: "As of 1840 credit expansion resumed in the United Kingdom and spread throughout France and the United States..."

Credit expansion, relative to the growth of base money.

The excerpt from de Soto includes this statement:

It is interesting to note that on July 19, 1844, under the auspices of Peel, England had adopted the Bank Charter Act, which represented the triumph of Ricardo’s Currency School and prohibited the issuance of bills not backed 100 percent by gold. Nevertheless this provision was not established in relation to deposits and loans, the volume of which increased five-fold in only two years, which explains the spread of speculation and the severity of the crisis which erupted in 1846.

The Act limited base money but not private-issue money.

The 100% backing established in Britain at that time did not apply to all forms of money creation. Therefore, it was a loophole that encouraged expansion of the unregulated forms. It encouraged the five-fold increase de Soto observed.

A five-fold increase in the monetary imbalance.

After the Bank Charter Act, only the Bank of England was authorized to issue bank notes -- to print money -- and those notes had to be backed 100% by gold. So printing money was strictly limited by the Act.

But the Act exempted demand deposits -- checking account money -- from the 100% backing, and left open the door to unlimited increase of private-issue money.


In the United States during the Free Banking era, the proliferation of private money typically ended not in inflation, but in crisis.

Economists today say that the business cycle was more frequent and more severe in the 19th century than it is today. They pat themselves on the back and attribute the difference to the wisdom of their policies. I say the difference arises from the fact that we are no longer on gold.

Policy had no control over gold. The growth of base money depended essentially on chance. Today the Fed can create a little inflation here and there...

I'm not saying it's the perfect policy. I'm pointing out the difference.

...and by doing so, relieve some internal pressures in the economy. The Fed's ability to increase the quantity of money works like the steam-release valve of a pressure-cooker. When we were on gold, the steam-release valve was stuck closed. Since inflation was not an option, the internal pressure had to find other outlets. Thus the more frequent and more severe crises of the 19th century.

Private Issues (2½): Imbalance and Reaction

It is the essential feature of banking, to generate money. And that is the origin of monetary imbalance. That is how M3 became so much bigger than base-money.

M3 and Base Money

M3, the blue line, increases until the red line (base money) can no longer resist and is pulled upward by it. There at the end you can see the big, sudden jump in base money, the Federal Reserve's reaction to a financial crisis that was caused by the gigantic long-term increase in private-issue money.

Monetary imbalance -- the excess of private-issue money relative to base -- is the recurring problem. It is the cause of inflation and of crisis and depression.