Sunday, July 31, 2011

Ouch, yuck, gasp, thud.

Catherine Rampell writes in the New York Times, Sure Cure for the Debt Problem: Economic Growth:

We face the largest budget deficit the nation has ever known: $1.6 trillion, the equivalent of about 11 percent of our economy. And, whatever Washington does, many economists say the situation will grow only worse...

But there is, in theory, a happy solution to our debt troubles. It’s called economic growth. No need to raise taxes or cut programs. Just get the economy growing the way it used to.

Good luck with that...

Good luck with that?? Catherine Rampell seems not to believe that the restoration of growth is possible, despite the title of her article.

“The basic issue is that the U.S. is on an unsustainable fiscal track,” says Dean Maki, the chief United States economist at Barclays Capital. “From that point, none of the choices are fun.” The most obvious choices, Mr. Maki says, are to reduce spending (ouch), raise taxes (yuck), let inflation run (gasp) or default (thud).

We wouldn’t need any of that if we could restore economic growth. If that happened, Americans would become richer and pay more taxes. Et voilĂ ! — we’d pay down the debt painlessly.

That is the only analysis of the problem in Rampell's article. Heck, it's the only one you'll find anywhere, except here at The New Arthurian Economics.

It is private debt, excessive private debt that hinders growth. Not public debt.

Since 1980 -- since Reagan -- we have been seriously trying to stop the growth of the Federal debt, with little or no success. Because it's the wrong debt.

After World War II, gross federal debt reached 122 percent of G.D.P., the highest ratio on record. But over the next 40 years, it fell to about 33 percent. That wasn’t because some blue-ribbon panel prescribed austerity; it was because the American economy became much, much richer.

The same happened during the prosperous 1990s, which began with deficits and ended with surpluses. Former President Bill Clinton is often credited for that turnabout, as he engineered higher tax rates. But most economists attribute the surplus years primarily to extraordinarily rapid growth.

After World War II, the Federal debt was large compared to private debt. Private debt was small and did not interfere with growth. So the economy grew. And over the next 40 years, private sector debt grew until the cost of it started to interfere with growth. Since then, economic growth has not been good... except briefly, in the late 1990s.

Something similar happened during the prosperous 1990s. In the late '80s and early '90s the rate of private debt growth fell substantially. This reduced costs and opened a window for growth. Then in the late '90s the economy grew enough to balance the Federal budget! But soon, the level of private debt was high again, and the economy was refusing to grow.

Congress has promised fiscal discipline for decades, only to undo its promises before the reckoning. Fiscal discipline is painful, unpopular and, in today’s Washington, as elusive as the prosperity we once had.

Congress has tried to achieve fiscal discipline for decades, without success. Why? Because reducing the Federal debt is the wrong approach.

We must reduce private sector debt, to free up the economy to grow again, and then the economy will grow.

When the economy grows, then we can balance the Federal budget.

"A version of the NY Times article appeared in print on July 31, 2011, on page BU1 of the New York edition with the headline: A Sure but Elusive Cure: Economic Growth."

Japan versus USA (3)

This graph shows the trends of M1 money growth for the US and Japan, each expressed as a multiple of its 1980 value. (In other words, I make both trend-lines start at the same level. This way, we can compare changes of the two lines.)

The blue line shows money in the U.S. There is a small hump in the money line around 1986, and a bigger one around 1992-1994. I have showed these two increases before.

In mine of 14 July, The Rise and Fall of the Non-Federal Relative, Graph #6 shows the two increases. Granted, discussing that graph I noted only the 1992-94 increase. But later, under Graph #9, I challenged you to investigate "two other times in the 1980s when the growth rate of M1 money exceeded the growth rate of total debt".

Three humps, total. If you look closely at the graph above, you may see a tiny hump between the years 1982 and 1984. That and the 1986 hump are the two I thought you might poke with a stick. The 1994 hump I've got covered.

Anyway, after 1994 Japan's M1 started growing faster, and U.S. M1 growth slowed. And you can see, after 2008 on the blue line, the ominous start of an increase similar to Japan's after 1994.

Japan versus USA (2)

The graphs from yesterday show GDP, total debt, and M1 money for the US and for Japan. I wanted to improve the comparison by putting GDP for the US and Japan together on one graph, total debt together on another, and M1 money together on a third. So that's what this post shows.

Ends up, I'm not very satisfied with how this post turned out. Thought about deleting it. Decided it might be useful to show that these comparisons are not satisfactory. But then, maybe you'll like'em.

GDP Comparison

The numbers on the vertical axis don't mean a lot. The units are "dollars divided by yen". And the GDP numbers are nominal, so inflation has a role in the trend as well. So, for what it's worth:

Where the trend shows flat, as from 1980 to 1991, the gist of it is that GDP growth in the U.S. and Japan was roughly equal. Where there is increase, the suggestion is that U.S. GDP was increasing more, or Japan's GDP was increasing less.

The two economies were growing roughly apace until 1991, when the U.S. economy started gaining on Japan's. As has been well documented, Japan's economic growth came to a sudden halt around that time.

Total Debt Comparison

Again, don't read too much into the vertical-scale numbers. But you can see from the flat trend (until 1990) that total debt in the two countries was increasing at about the same rate. As with GDP. And as with GDP, after 1990 total debt in the U.S. started growing consistently faster than in Japan. Again, it is because of the slowdown in Japan.

The last two bars on the right in this graph are again about the same height. If that pattern continues and becomes a trend, it will be because total debt in the two countries is again growing at about the same pace. Or, perhaps I should say, *not* growing.

M1 Money Comparison

More variations in the trend here. This graph is not so easy to read. For context, look again at the three trend-lines: Total debt, M1 money, and GDP for Japan, 1980-2009:

Things were "normal" in Japan until the late 1980s when the pace of debt growth suddenly slowed. Then GDP flatlined, and then the quantity of money began abnormal increase. That abnormal increase is reflected on the bar graph above, as a decline since the mid 1990s. At that point, Japan's M1 money started growing significantly faster than U.S. M1 money.

Saturday, July 30, 2011

Japan versus USA

Graph #1, some numbers for Japan, from mine of 29 July:

Graph #1: Money, Debt, and Output in Japan

Graph #2 for today, showing comparable numbers from FRED for the USA:

Graph #2: Money, Debt, and Output in the USA

In both countries, debt is the big number. But look how small the U.S. M1 number is, compared to Japan!

Also, not to make predictions here, but on Graph #2, toward the right end, the blue line looks an awful lot like the red line from Graph #1 from 1980 to 1992: a consistent uptrend and then a sudden flattening. One hopes that the flatness of our GDP does not last as long as Japan's.

Come to think of it, U.S. debt (Graph #2, red line) from 2000 to 2010 looks very much like Japan's debt (Graph #1, yellow line) from 1980 to 1990. And I should point out that after 1990, Japan's debt continued to increase.

That's the problem right there, I think.If you want to fix the economic problem by inflating the money supply rather than by crashing the debt, it is a long slow process. The inflation has to work its way through the system, bringing all prices up with it, and nobody likes inflation. Moreover, the money and the prices have to go up faster than the debt is going up, or it is all for naught. As you can see, in Japan the debt continued to increase. Not fast enough to create a vigorous economy, but fast enough that debt remains a growing problem.

If you just crash the debt, you don't have to have inflation. You can maybe bring debt back down in line with everything else.

Friday, July 29, 2011

Japan (7): Three Trend Lines

A picture of the raw numbers gives a feel for what has happened in Japan:

Graph #1

Same data, showing percent change from previous year:

Graph #2

Graph #1 shows GDP flat since 1990 and M1 money climbing to meet it, while total debt sits in a holding pattern at over five times the level of GDP.

Graph #2 shows GDP settling down around zero percent growth by 1993, and debt settling down to zero growth by the latter 1990s. These observations are reflected in Graph #1 as horizontalness of the red and yellow trend lines. The only growth shown on Graph #2 since the early 1990s is the growth of the money supply.

Graph #1 again: The red line, GDP, goes flat. But the plan is that GDP should get bigger: Make the pie bigger so everyone can have more. So the flatline is no good.

Graph #1, the blue line, M1 money, goes up. If you look at the space between the blue and yellow lines, that is a pretty big gap. The yellow line, we can safely assume, is not money that people earned and spent, but money they borrowed and spent, people and businesses and government. The blue line shows how much money there is in the economy that can be used to pay off debt.

Actually, even the blue line includes some debt. So there is even less money than the blue line shows, that can be used to pay off debt. And the blue is far below the yellow.

Now, the blue line doesn't have to be as high up as the yellow line, because money changes hands. I might bring the yellow line down by paying off my debt to you. And then you might bring it down more, using the same money to pay off a debt to someone else. The money gets recycled. So it is okay if the yellow line is somewhat higher than the blue.

But the gap between the yellow and blue lines is filled with cost. It is filled with the cost of interest. And the bigger the gap, the bigger the cost. That cost is an obstacle standing in the way of productive work. That cost is an obstacle that interferes with the growth of GDP. It is an obstacle that consumes money that should be going to wages. It consumes money that should be going to profit.

It is true, as people often say, that money paid as interest expense is also interest income. But where is the output in that? Where is the product? Where is the GDP? Where is the stuff produced? There is no product to offset the cost. And that is the trouble with excessive debt in Japan, in the U.S. and everywhere.

Previous JAPAN post:  7-25-2011

Thursday, July 28, 2011

The best Billy's got?

From today's Billy Blog:

If you think about it for one second or less you will realise this is a totally fraudulent representation of the situation in the US. The “red ink” jargon is accounting and auditing terminology which refers typically to the recording of a financial loss or revenue deficit for a private firm. The implication is that insolvency is possible.

There is no application of that concept to a sovereign government – one that issues its own currency...

That's it. That's all Billy has to offer. Insolvency is not possible if you can print your own money.

Billy, maybe you are right. But it is a damn weak argument.

Glance at the relation between Federal debt and everybody else's debt in the U.S. over the past several decades. The economy does well when our debt is growing, relative to the Federal debt.

The economy does poorly when private debt so vast that it impedes its own growth.

Billy, this is a good argument.

Real Time Analysis

I will use a relevant measure of Federal debt. I don't know why everybody talks about the $14.3 trillion debt when the relevant measure is less than $10 trillion, but there must be reasons. Anyway, I already did my debt graphs with the $14.3 debt series. Now I'll use the other one.

But first...

Jim suggests that I use FGSDODNS as a measure of the Federal debt. That's "Debt Outstanding Domestic Nonfinancial Sectors - Federal Government Sector". It looks like this:

But I'm thinking about using the one that comes up when search FRED for "Total Credit Market Debt". That's "Total Credit Market Debt Owed by Domestic Nonfinancial Sectors - Federal Government" (FGTCMDODNS).

That Series ID is not too hard to read, by the way (I keep telling myself):

  • FG is Federal Government,
  • TCMDO is TCMDO (Total Credit Market Debt Owed), and
  • DNS is Domestic Nonfinancial Sector (of which FG is part).

So how does the "Debt Outstanding" total compare to "Total  Credit Market Debt"?

Pretty close. (There are only a few blue dots peeking through the red trend  line.)

I wonder how they look in a ratio, showing Jim's number as a percent of mine?

Weird, but close. Always within two or three percent, in a pattern that looks like the result of some bookkeeping rule or some bureaucratic imposition.

But it gets me thinking. I use TCMDO because I found TCMDO. I'm gonna search FRED for "Debt Outstanding" and see what I get. Maybe there's better information under that name...

24 items, including the BofAML stuff again, along with some discontinued series. Try "Debt Outstanding Domestic" instead?

22 items. And it probably dropped  yadda yadda Foreign, which I want. I'll go with the 24 items.

Looks like there are three categories to add together to get "total" debt outstanding. Just like I did yesterday for components of TCMDO. But I don't see a total, comparable to TCMDO. That's okay. I know how to make one. I'll add these three together:

  • Debt Outstanding Domestic Financial Sectors (DODFS)
  • Debt Outstanding Domestic Nonfinancial Sectors - Total Domestic Nonfinancial Sectors (TODNS)
  • Debt Outstanding Foreign Financial Sector, Rest of the World Credit Market Instruments, Excluding Corporate Equities Liability (DODFFSWCMI)

and them compare the total to TCMDO...

Yep, the three added together peak at about 52 thousand billion, same as TCMDO. And with the latter overlaid on the total in red...

Call them equal.

Good. So I can use my TCMDO numbers, and not waste the shopping experience.

Wednesday, July 27, 2011

Bagging the Groceries

Looking at what I have to work with here, in the FRED debt data on my Shopping List (see previous post).

Debt owed by Domestic Nonfinancial sectors, plus debt owed by Domestic Financial sectors, plus "Foreign Sectors - Rest of the World..." add up to a red line that completely covers the blue TCMDO (total debt) line on this graph:

So that's a match. Foreign is a very small part of it, by the way.

Next: Find the components of the Domestic Nonfinancial debt.

Following the items on my shopping list, I include

  • Household debt
  • Federal Government
  • State and Local Governments
  • Farm Business
  • Nonfarm Noncorporate Business
  • Nonfinancial Corporate Business

Once again, remarkably, the components added together in the red line completely cover the blue line. The six components together equal total domestic nonfinancial debt.

Another match. So far, so good.

Next: Find the components of the Domestic Financial debt.

The shopping list includes "Commercial Banking" and what look to be three sub-sectors of commercial banking. I will test that assumption and if it holds, eliminate the sub-sectors from the components of Domestic Financial debt.

The sub-sectors are

  • U.S.-Chartered Commercial Banks
  • Bank Holding Companies
  • Foreign Banking Offices in U.S.

With just the first two of these, the red line over-writes the blue for the recent years (since 2000) and also before 1980 (though in close-up, it might not).

The three components added together do completely cover the blue line:

Back to the components of Domestic Financial. Following the items on my shopping list, I include

  • Agency- and GSE-backed Mortgage Pools
  • Government-Sponsored Enterprises
  • REITs
  • ABS Issuers
  • Commercial Banking

With these five in place, the red line has taken on the shape of the blue. But it peaks at $14 trillion, where the blue peaks at $17 trillion. So I keep going.

  • Funding Corporations
  • Brokers and Dealers
  • Saving Institutions
  • Finance Companies
  • Life Insurance Companies
  • Credit Unions

After adding these in, down to Finance Companies, the red line almost entirely over-writes the blue:

But the graph title and the key don't handle the long description well :)

With the next one I hit a brick wall. FRED reports:

Line 2 has the maximum of 10 data series added. You must remove a data series from this line before you can add another.

So I back up, and add a third line. I'll put the last two items into that line and if it rises negligibly above zero, I'll figure I have determined that there are eleven components of FRED's Domestic Financial debt.

Close enough.

And that's the Sunday morning exercise.

Tuesday, July 26, 2011

A Shopping List

At FRED, in the search box I type TOTAL CREDIT MARKET DEBT and press ENTER.

I get a list of 32 data series (as of 16 July 2011) including TCMDO, debt owed.

There is also debt owed by Domestic Nonfinancial sectors, and debt owed by Domestic Financial sectors.

I still think it emphasizes our flawed focus, that these two sectors are called financial and non-financial, rather than financial and productive, or maybe non-productive and productive. I think it says a lot about the way we look at things and what's wrong with our economy.

In addition to Domestic Financial and Domestic Nonfinancial, there is one non-domestic category: Foreign Sectors - Rest of the World Credit Market Instruments, Excluding Corporate Equities Liability.

The domestic categories are subdivided. Under Domestic Nonfinancial one finds

  • Household sector
  • Federal Government
  • State and Local Governments
  • Farm Business
  • Nonfarm Noncorporate Business
  • Nonfinancial Corporate Business

Under Domestic Financial one finds

  • Agency- and GSE-backed Mortgage Pools
  • Government-Sponsored Enterprises
  • REITs
  • ABS Issuers
  • Commercial Banking

    • Commercial Banking, U.S.-Chartered Commercial Banks
    • Commercial Banking, Bank Holding Companies
    • Commercial Banking, Foreign Banking Offices in U.S.
  • Funding Corporations
  • Brokers and Dealers
  • Saving Institutions
  • Finance Companies
  • Life Insurance Companies
  • Credit Unions

In addition to these, there are BofAML categories. All of these categories end with the phrase "Plus Sub-Index Total Return Index Value", which I have removed (after having typed them all in) to make the list more readable:

  • Public Sector Issuers Emerging Markets Corporate
  • Private Sector Issuers Emerging Markets Corporate
  • Non-Financial Emerging Markets Corporate
  • Financial Emerging Markets Corporate
  • Public Sector Issuers US Emerging Markets Liquid Corporate
  • Private Sector Issuers US Emerging Markets Liquid Corporate
  • Non-Financial US Emerging Markets Liquid Corporate
  • Financial US Emerging Markets Liquid Corporate

But these all turn out to be some "index" measure. Not some portion of total debt.

Just your typical Saturday morning exercise.

Monday, July 25, 2011

Japan (6): I like that Cabinet Office

I like the Cabinet Office. A good source of economic statistics on Japan.

I pulled "21a1_en.xls" for Gross Domestic Product Account (Production and Expenditure Approach)(Excel:52KB)

I thought for a minute I might be able to compare the approaches to GDP-counting, as others have compared the "product" and "income" approaches for U.S. data. But the two GDP totals are equal in the Japan data, so I set that notion aside.

So then I just did my usual dollars-of-money-in-circulation compared to dollar-value-of-output graph, this time for the Japanese economy... in Yen:

Starts out low, ends up high.

Starts out low, but not by U.S. standards.Through the 1980s, Japan had 30 cents for every dollar's worth of output... or, 0.30 Yen for every Yen's worth. Anyway, it is something like three times what we had in the U.S. by the end.

Not that that matters for Japan. It shows that the U.S. number was really, objectively low. But for Japan, for both nations, what matters is continuity and context. Can't tell from the graph where things were before 1980. But given what the graph does show, Japan's M1 number was low in the 1980s.

But that started changing after 1990, with the money supply going uphill at a pretty good clip. There is a big jump between 2001 and 2002. The timing of this corresponds to the increase in base money Krugman shows. (See the two Krugman graphs in my previous post.)

By 2009, amazingly, the quantity of money in circulation is more than the GDP. It means money is hardly moving at all over there.

It also means I have to think about things some more. For thirty years I've been saying we need to increase the quantity of money in circulation (relative to nominal GDP). But I just wanted to get it back up from 10 or 12 cents, to 25 cents or so. In Japan, they now have the equivalent of a dollar -- a dollar in somebody's pocket for every dollar's worth of stuff they produce in a year. That's a *lot* of money, twice what the U.S. had at the end of World War II.

In my experience, M1 circulates. I get paid, and the money goes quickly. I'm lucky to have anything left when the next paycheck comes around.

So to my way of thinking, an increase of money in circulation means I have a chance to grab more of it. We have a chance to grab more of it. Maybe I can get a raise. Maybe business is expanding. Maybe more people can get jobs. Maybe things pick up some. Because the money is circulating.

And yeah, maybe prices go up some. But I offset that with tax incentives to accelerate the repayment of debt, which slows the growth of credit-use, which compensates for the increase in the money supply. It's a trade-off as far as inflation, except that the money itself becomes less costly to use. That eases inflationary pressures, so there is a net gain. And, looking at the history of money and credit, looking at my U.S. DPD graph, there can be no doubt that credit-use made using money expensive.

But my experience is with a growing economy, or at least a pre-crisis economy. Over at Twenty-Cent Paradigms, Bill quotes Ezra Klein:

Keynes — and others who later elaborated on his work, like Hyman Minsky — taught us that although markets are usually self-correcting, they occasionally enter destructive feedback loops in which a shock to, say, the financial system scares business and consumers so badly that they hoard money, which worsens the damage to the system, which further persuades other economic players to hoard, and so on and so forth.

That's what happened. We are now in the post-crisis economy, as is Japan. No one any longer thinks what I think: that we can grow the economy. I think if we caught it early, we might have been all right. Ah, but there is no longer the urgency. Now there is the idiocy instead. Debt-ceiling, idiot-versus-idiot idiocy.

Sorry 'bout that. Anyway, in Japan they have more money than output, and still their economy does not grow...

If they have that much money in Japan and still their economy does not grow, then I'm wondering where the money is. If it is "circulating" then how can it be that the economy is *not* growing? They don't even have inflation over there!

So I'm thinking, maybe the money is all bunched up in too few hands.

For the U.S., based on my experience, I have always thought that if we had an increase in circulating money, it would be pretty well distributed, by default. The invisible hand, and all. Shipping containers full of hundred dollar bills is not really part of my experience.

But maybe it is necessary to force some distribution of the money by policy. Maybe only because we waited too long, or maybe it is the nature of the beast. I have to think about it more.

Or maybe, money is bunched up in too many hands. Look at me. Since the coffee-shop closed for lack of business, I've been saving my coffee money. Well, not saving. I stick small bunches under the mattress you know.

Next JAPAN post:  7-29-2011

Japan (5): Debt per Yen

A version of my "Debt per Dollar" graph, for Japan's economy:

Rise to 1990. Decline to 2002, Then very slow decline, almost flat. The trend line of the graph is comparable to Japan's economic performance, given that "the lost decade" began in the early 1990s and continues today.

Total debt in Japan has fallen by roughly two-thirds, relative to Japan's M1 money.

Wikipedia (20 July 2011) says the Bank of Japan defines M1 as "cash currency in circulation + deposit money". That's comparable in my book to what the St. Louis Fed says for the U.S. definition:

"M1 consists of: (1) currency outside the U.S. Treasury, Federal Reserve Banks, and the vaults of depository institutions; (2) traveler's checks of nonbank issuers; (3) demand deposits; and (4) other checkable deposits..."

So I think my use of M1 in a similar way for graphs of U.S. and Japanese data is satisfactory.

The debt numbers come from hbl at Thought Offerings, updated using currently available info from the Cabinet Office, as discussed earlier in this series of posts. The M1 numbers come from EconStats. (Annual data.)

What is the significance of the graph? It is too soon for me to say. But while looking for references to the phrase "the lost decade" I found two related graphs in Krugman's posts:

Source: Paul Krugman

The Bank of Japan accelerated the growth of base money during the 2000s, as the Federal Reserve did in the early 1990s. (I draw no conclusion at this time.)

Source: Paul Krugman

The rapid acceleration of base money had no noticeable effect on  M2 money... And perhaps little effect on M1 money, if Japan's economy is anything like ours. My first graph here shows that a large drop in debt relative to base money (caused by a large increase in base money that you likely remember) had no effect on MZM (roughly equal to M2) and little effect on M1 money.

(I draw no conclusion at this time.)

// Update before posting.

I have been looking at this and other graphs of Japan data for a week now, and I am ready to offer a conclusion. The debt-per-dollar ratio (debt-per-yen in this case) has fallen substantially in the past 20 years. Low as it is, however, Japan's economy seems not yet to have picked up the pace of growth.

I conclude that the DPD is an excellent pre-crisis indicator, but it may be ineffective after the crisis.

I observe that Japan's DPD has fallen not because debt fell, but because the quantity of M1 money was pushed up to an absurd level -- greater even than the GDP!

I conclude that the necessary action is the reduction of debt.

So much is beyond question.

Furthermore, I want to conclude that it is private-sector debt which must be reduced. However, I may not have seen Japan graphs that make this conclusion inescapable.

Sunday, July 24, 2011

Japan (4): An Asian Relative

At last:

Japan's equivalent of the Non-Federal Relative: from over 6 to less than 2
This graph lacks the "random walk" appearance of the U.S. debt-relative. Rapid rise to a (rounded) peak, followed by an interminable decline that has been euphemistically called a lost decade.

By no coincidence, the trend line peaks in 1989-1990, and Japan has been in its grand funk ever since.

This graph is hard for me to read because I don't know the highlights of Japanese economic history. Did they have a "golden age"? Did they have a "macroeconomic miracle" like ours of the latter 1990s? Did they have a "Reaganomics"? I don't know. So my analysis of the graph is somewhat limited.

And then, there's no hint or wiggle of recession anywhere on the graph. Maybe that indicates real structural differences between the Japanese and U.S. economies. Or maybe it indicates that the debt numbers are not to be trusted. (For now, I'm assuming the numbers are good and the structural differences are real. And I'm not one to use the word "structural" lightly.)

But clearly, like the U.S., in Japan they use credit for growth. The Non-government debt-relative increases during the boom -- until around 1990 -- and decreases in the decline.

Two questions occur to me.

1. How does what this graph shows fit with what I said before about the debt-relative?

The trend line rises while the economy is good, and declines while the economy is not good. In addition, a high level of the debt-relative is correlated with the onset of hard times, as it is in the U.S. economy.

2. What do I have to reconsider?

It seems that the non-government debt relative can go quite low, and that the low level alone is insufficient to create an economy bursting with vitality.

But I have a thought about that. Maybe the quantity of money in circulation is still low in Japan, and this is preventing recovery. Remember, in mine of 14 July I showed that the "macroeconomic miracle" occurred after two things happened: There was an unusual slowdown in the growth rate of debt. And there was an unusual acceleration in the growth rate of money in circulation.

Saturday, July 23, 2011

Japan (3): Checking my Numbers

Source: Thought Offerings

The graph above is from hbl's post. It covers the years 1980 to 2007.

The graph below is my attempt to duplicate hbl's graph using the Excel file I talked about in the previous post. It covers 1980 to 2009.

Source: My Reconstruction of the Numbers

Wow. hbl's is prettier. Oh, well.

My intent is to see if my graph shows the same general trends as hbl's. It seems to be a pretty good duplicate. I take that to mean I followed the steps correctly, gathering and handling the data.

Looking at just the uppermost trend line, total debt, I do see a few little high spots that don't appear on hbl's graph: 1997, 2000, 2007. So I checked those and the adjacent years: compared the numbers I used for the graph to the original numbers from the original download file. Same numbers. I didn't glitch it. Probably the numbers just got updated since hbl did his study.

So now I have some Japan debt numbers I can use!

The result of my efforts is available at Google Docs in two forms:

The graphs are imperfect in the browse file -- no labels on the key (which kind of defeats the purpose!). The Excel file was created in OpenOffice.

Friday, July 22, 2011

Japan (2): A Long Hard Slog

This post will be useful if you want to find the data on Japanese debt. Otherwise, you might want to skip it... unless you want a lesson in how difficult it is to use (or to explain) an inconsistent internet interface.

At Thought Offerings, in hbl's post The Mystery of Japan's Private Debt Levels (Solved?), in the Methodology for Obtaining Japan's Debt Data section, the link to "the aggregate national accounts stock data" is broken. But the error page provides a link to the Cabinet Office Home Page.

At the home page, under "Statistics" is ESRI(GDP, Business statistics). Clicking ESRI brings you to a page with a few bold headings, including

  • What's New (ESRI)
  • Monthly Economic Report and Other Reports
  • Business Statistics (ESRI)
  • SNA(National Accounts of Japan) (ESRI)

Under "SNA(National Accounts of Japan)" the second item is Annual Report on National Accounts. Clicking that turns up three options:

  • National Accounts for 2009
  • Information of corrections
  • Release Archive

Clicking National Accounts for 2009 brings up a page that offers three options across the top of the page: "Flow", "Stock", and "Trial Estimation". Clicking Stock brings up a categorized list of Excel files. Here, I'll show you the part of it you need:

Among the supporting tables they have Closing Stocks of Financial Assets/Liabilities : All Main Sectors: Stocks of liabilities (Excel:447KB)

Click it, and you download an excel file. Turned out, that was the one I needed. The name of the file I got was "21ss62_en.xls".

Now... hbl provides a description of what to look for in the Excel file:

What I found for 2009 is a pretty good match to this, close enough to confirm I'm on the right track. hbl uses two subtotals to count as debt: "Loans" and "Securities other than shares". In the post he's not completely confident that his selection is the best match to comparable U.S. data. I'm not, either, but it seems right to me. Anyway, I'll select the same rows, and have high confidence that my graphs will be comparable with his.

The Excel file provides a separate worksheet for each year from 1980 to 2009. So I have to pull out the numbers I need and compile them myself. But that's okay. At least I have numbers to work with!

Thursday, July 21, 2011


I spent an hour googling for statistics on total debt in Japan. Found two potentially useful sites: The Statistics Bureau, and The Bank of Japan. But no useful information until I got to Thought Offerings: The Mystery of Japan's Private Debt Levels (Solved?)

Right off the bat, Thought Offerings' host hbl offers a graph showing "Japan's Total Debt Levels by Sector (1980-2007)". Total debt, plus a breakdown into broad categories. Of this chart, hbl writes:

The above chart shows the total debt in yen for Japan's household, corporate, financial, and government sectors from 1980 to 2007 (the latest available). Private sector debt actually kept rising (much more slowly) after 1990, peaking around 1997! Government debt also rose significantly during this time period.

Two remarks on that excerpt. First, hbl breaks debt down the way I do: private sector debt and government debt. (And the highlighting is hbl's, not mine.)

Second, Japan's economy got into trouble around 1990, but private sector debt continued to grow for almost a decade after that. If (as I think) private sector debt is the problem, then the problem continued to get worse for a decade after it became obvious. There is a lesson in this for U.S. policymakers: If we expand government debt in order to rescue the economy, we will continue to make the problem worse unless we take specific actions to reduce private-sector debt as part of the rescue plan.

As a follow-up to that second remark, hbl again:

... willingness to provide ongoing fiscal stimulus as Japan did ... makes the Japanese style stagnation and mild deflation relatively more likely in the US

More likely than a deep depression, hbl says. But surely, purposeful expansion of the government debt does not solve the problem. What we need instead is purposeful reduction of private debt. Because excessive private debt is the problem.

I hope you can see, in the above, my respect for the work hbl has done in the post. I find a lot of insight in the post, and a lot to agree with. Now I'm going to go the other way and point out what I see as the major flaw.

I do agree with this: "Of course, total debt-to-GDP is not the only macroeconomic determinant that matters..." But these are the "other factors" that hbl identifies:

  • Substantially higher household debt (in the U.S.)
  • The global context today, and
  • Other differences in financial markets today

Hbl follows that analysis with this insight:

Many people summarize the options for removing excessive debt as inflate or default.

The options for removing excessive debt. That is what needs to be done. That must be the primary focus. Nothing else will solve the problem. hbl knows it, but has difficulty maintaining focus. hbl's concerns include household debt, and the global context, and financial markets. A piece of the problem, a result of the problem, and a vehicle delivering the problem.

But where is the focus on "removing excessive debt"?

And where is the "macro-economic determinant that matters"? Where is the analysis of the relation between private debt and government debt? Where is the debt-relative? Everyone focuses on government debt and if they ever compare it to anything, they compare it to GDP. Allow me to summarize the evidence hbl presents for Japan:

  • The graph of total debt levels by sector
  • A graph of Japan's GDP.
  • A graph of the debt-to-GDP ratio by sector
  • Another graph of public and private debt as a percent of GDP
  • A comparison to U.S. debt-to-GDP by sector
  • A bar graph of debt-to-GDP ratios
  • A table of percentages related to the bar graph

Of those seven items, one shows debt, one shows GDP, and five show debt-to-GDP. And yet, as hbl says, debt-to-GDP is not the only macroeconomic determinant that matters. So where is the debt-relative?

hbl "breaks debt down the way I do: private sector debt and government debt." But hbl doesn't use his own breakdown. He doesn't compare the one to the other in his graphs. He doesn't show private debt relative to government debt.

I want to see some graphs comparing private debt to government debt in Japan. That is what's missing from hbl's post.

Where is the debt-relative?

Wednesday, July 20, 2011


The vocal majority today looks at the federal debt and deficits and says This must stop. It looks at medicare and baby-boom demographics and says This must not continue. It looks at what's happening in America today, and says This must change.

The vocal majority is looking at consequences of the economic problem, and it is ready to change those consequences.

Unfortunately, it thinks it can change the consequences directly, without changing the underlying problem. That is a mistake.

Actually, there is a relevant statement I want to quote:

In this lies almost the whole difference between good economics and bad. The bad economist sees only what immediately strikes the eye; the good economist also looks beyond.
From Economics in One Lesson by Henry Hazlitt

Well, here's a fuzzy

A few days back, I was looking at Non-Federal debt relative to Federal debt, and Jim thought I was using the wrong Federal debt number.

Yes, I'm still dwelling on it :)

The difference between the two Federal debt numbers is that one includes, and the other excludes, the debt held internally by the government. The bigger number, which includes internally-held debt, is the $14.3 trillion that everybody talks about. The smaller number, under $10 trillion, excludes the internally-held debt.

So, what's the difference? The graph below shows the bigger number as a multiple of the smaller:

Until 1974, and again since the 1980s, internally held debt was increasing relative to externally ("publicly") held debt. Most significant is the increase during the 1990s.

The 1990s. In the early years there was a sustained increase in both base money and M1 money. In the latter years there was a burst of growth.

I'm wondering how this graph fits into that picture.

The Federal Relative and "relative to GDP"

This is the Federal Relative; we looked at it recently:

Graph #1: The Federal Relative

Since (as you can see) the Federal Relative is so low, I woke up today wondering what is it that drives people crazy about the Federal debt. Maybe it bothers people that the debt is so big, relative to GDP:

Graph #2: The Federal Relative and Federal debt relative to GDP

The red line here shows the Federal debt, relative to GDP. It's the "size" of the Federal debt. It's big. Since the mid-1960s it has been bigger than the Federal Relative. And though they show similar patterns since that time, the red line has been pulling away, rising higher and higher. And maybe that's what bothers people.

Still... to drive the red line down, we must reduce the Federal debt relative to GDP. But if we want to reduce the Federal debt, we must also push the blue line down. In order to push the red line down, we must push the blue line down. And this is a problem, for the blue line is already just about as low as it can go.

When we take Graph #2 and add to it (in green, below) the trend line of Non-Federal debt, relative to GDP, we start to see what a high level of debt really is:

Graph #3: also comparing Non-Federal debt to GDP

If you get the feeling it is the Non-Federal debt that's pulling everything else up, I think you're right.

Add the Non-Federal Relative to the graph, and it stands out above all:

Graph #4: with the Non-Federal Relative added in
Once again, the Non-Federal component of debt is by far the biggest part of the debt problem. Everything else pales beside it. (And this graph shows the understated version of the Non-Federal Relative.)

On Graph #4 here, in the red and blue lines, the Federal trends, you can see slight increase through the 1980s and into the 1990s.

In the Non-Federal trends, you can see a flattening in the green line, mid-1980s to mid-90s. And in the orange line you can see that flattening translated into a brief but significant decline, 1989-1994. That came just before the vigorous uptrend that accompanied our best recent years.

Again: It was the fall of the orange line, Non-Federal debt, that allowed our economy to perform well, however briefly.

What we need to do is bring the Non-Federal Relative down to perhaps half where it has been for 30 years or more.

The best way to do that? Reduce private-sector debt, the green line. Not the red and blue.

And yes, very often I get up at two in the morning, because my restless mind put two notions together and I want to see what the graph looks like.

Tuesday, July 19, 2011

Coin of the Realm

Anybody else having trouble with the post-scheduler in the new blogger interface?

An interesting development: Supposedly, the U.S. Treasury could have a couple trillion-dollar coins minted on the cheap, and deposit them at the Federal Reserve. That would increase the balance in the Treasury's checking account. It would let the Federal government get around the debt ceiling limit -- because the coins are not debt.

Most interesting.

But anyway, context: The idea is spreading on MMT blogs, so I take it for an MMT idea. And therefore I must bring up another MMT idea, one that disturbs me immensely: the notion that "all money is debt".

Here's the rub: The beowulf coin is a way to get around the debt limit because the coin is not debt. But the people most in love with the coin are the same people who go around saying "all money is debt". So I have to ask, is the coin money? Because if it is money, then it must be debt. That, or the "all money is debt" thing is wrong.

The "all money is debt" idea is some kind of philosophical definition, I think. It is not an economic definition. To me, to most people, a debt is something that must be repaid. Money that you borrow, for example, must be repaid, which is why the lender keeps track of it. And the amount the lender tracks is a debt.

But if you or I or the Federal government happens to have a dollar, that dollar is not debt. Money that you earn is an example of money you do not have to pay back. You are not obligated to "pay it back" to anybody simply because you have it.

I object to the "all money is debt" notion for one reason: it blurs an important distinction. It says that money you have to pay back, and money you don't have to pay back, are exactly the same.

Arthurian economics is based on the distinction between money and debt, and on the increasing cost of money in the economy as a whole, as usage shifts away from money you don't have to pay back, to money you do have to pay back.

But it is much easier to understand what I have to say if you allow me to use the word "money" to mean I don't have to pay it back, and the word "debt" to mean I do have to pay it back. Then it becomes quite simple.