Some key statistics as prediction aids

M3b FTTM International dollar flows Trade & budget Yield curves Most of world GDP & money GDP & money creation US credit "Total money" Global liquidity Hard vs. paper assets Money supply link to inflation Pain misery index L money supply

Money supply M1, M2, M3TMS, Debt & Credit
M1M2M3 Credit Debt TMS

CPI and CPPI Velocity

UnemploymentTotal Money Supply

Total credit, with sectorsMoney supply values
Credit breakdown M1M2M3actual

Total money supply, w/deriv.Austrian AMS (TMS1) weekly

M3 is back

We did some sleuthing and data extraction and put M3 back together from various weekly Federal Reserve reports that are still available.
  1. The formula we're using has five 9s correlation to the original data back to 1980.
  2. There is only one missing element that is apparently no longer available (Eurodollars) and an adjustment has been applied to generate it. Its only about 3% of total M3 so should not have a material effect on the total.
Here is our article on M3b, which details our work and notes the sources for the data. Note that as of Nov. 10, 2006 the Eurodollar estimation formula has changed - see the article for details.

John Williams monthly reconstruction of M3 is here. Ours tends to be more volatile than his, partly because it's weekly and partly because of our differences in calculating the repo and Eurodollar component of M3.

Finally and to put M3 into proper perspective with inflation (as measured by CPPI), the M3 and M2 strong inflation link is virtually unquestionable. The longer term inflation picture is clear, although M2 shows a pause and likely temporary disinflation as of 2008. Certain bloggers are incorrect and have continually avoided these facts and the linked chart.


M3, longer term chart


Same, log chart

M3 consists of M2, institutional money market mutual funds, time deposits in amounts of $100,000 or more, repurchase agreement liabilities of depository institutions (in denominations of $100,000 or more) on U.S. government and federal agency securities, and Eurodollars.

For reference, and as of early 2007, M3 is about $11.5 trillion, M2 about $7.1 trillion, institutional money markets funds about $1.4 trillion, jumbo CDs about $1.7 trillion, repos about $.67 trillion ($670 billion) and estimated Eurodollars are about $.61 trillion ($610 billion).
11/30/2007 - Note that much of the large growth in M3 lately has been in flows into CDs and Money Market Funds, a normal occurrence during financial turmoil. See our financial crisis page for more detail, and a picture of the current level of a U.S. financial crisis.

3/31/2008 - Also see our best effort construction of all Fed activities here. As of today, it shows that the Fed has been backing off on total money creation activities on a relative basis since about December 2007.

4/4/2008 - As of 3/19/2008, we have added the results of the new Fed TAF, TSLF and PDCF "tools" to our M3 reconstruction, since they are quite similar to temporary repos (repurchase agreements). Temp repos are part of the original definition of M3. Note that they add very roughly about 2-2.5% to the annual growth rate.

5/30/08 - Removed TAF, TSLF, etc. from the "regular" M3 weekly chart and now maintaining it as just a link here (M3 with TAF, etc.).

Important change - December 7, 2008

An adjustment has been made as of this week to account for changes in the Eurodollar component of M3, taking both derivatives and the flight to the dollar into account. It was back dated to September 2008.

Important addition - December 15, 2009

12/15/09 - added the rest of the Fed temporary and permanent OMO-like repo programs into our alternate M3 reconstruction here. It now includes TAF, TSLF, PDCF, borrowed reserves, MBS/GSE puchases, Supplemental Financing Program borrowings, adjustments for dollar swap lines, etc. This chart is an attempt to follow the actual original definition of M3 (which includes "repurchase agreement liabilities of depository institutions"), and update the M3 data taking the huge Fed policy changes since 2007 into account. Note also that the dollar swap line adjustment is minor, since the Treasury's S.F.P. program provided most of the financing.

Finally and to put M3 into proper perspective with inflation (as measured by CPPI from our work), the M3 and M2 strong inflation link is virtually unquestionable. The longer term inflation picture is clear, although M2 shows a pause and likely temporary disinflation as of 2008. Certain bloggers are incorrect and have continually avoided these facts and the linked chart.

M3 versus MZM, short term chart showing the high similarities
M3 versus MZM, long term chart showing the similarities

M3b, with the effects from the TAF, TSLF, PDCF etc. included
M3b, with the effects from the TAF, TSLF, PDCF etc. included (long term)

M3b, with 13 week rate of change
M3b, with 13 week rate of change (long term)

M3 raw data in Excel, weekly since 1900 (interpolated up to 1959), updated monthly
Same, except zipped

Fed & Treasury total money supply

All major Fed operations (plus custodials), showing the running total of all Fed & Treasury controlled money creation or destruction actions.


The red line shows the smoothed (via 13 week moving average) annual change rate of all actions, and the black line adds in trading information of the Fed's primary dealers (GSDS). The blue line is the same as the red line, except is not smoothed with a 13 week moving average.

Please see the glossary for all definitions ("helo" drops defined as gov't tax rebates & stimulus checks, etc.)
Source for much of data

Same data, long term

Fed all, long term

Same data, short term, different format

FTTM mod1

L money supply, reconstructed



International dollar value and trend

The quick explanation: TIC (Treasury International Capital) flows, when the trade and budget deficits are subtracted, are nicely correlated to the international value of the dollar and show the trend and expected trend quite well.

The in depth explanation: Click here for a full definition of TIC (Treasury International Capital) flows. Basically and simply though, its money that foreign banks and other official institutions flow to the US - its their investments in US bonds and stocks, and is measured in billions of dollars every month.

Two other factors are included in the chart - the US budget deficit (how much more money is being spent than being collected by the US government), and the US trade deficit (how much more money is being spent on stuff being imported into the US from other countries than the US is exporting to other countries).

In order to value a business or company and see what it's worth and how it will do in the future, three of the most important factors are sales, expenses and profit. Profit is basically sales minus expenses. If we go way out there and assume the entire US is a company, and pretend that the US dollar is its stock, then we have another way to look at the international value of the dollar.
The black line on the chart below is the monthly TIC flow from other countries (income), with both the trade and budget deficits (expenses) subtracted. So you say - "So what?"... well, by doing that we show an income and expense statement for the US dollar itself. Any numbers above zero on the left hand scale mean a profit and if the number comes in below zero then there has been a loss.

In other words, if we back way off from the dollar and look at it from a 30,000 foot level as the stock of the USA itself, we need to figure out what would represent sales and what would represent expenses. We pretend that TIC flows are income and that the combination of the trade and budget deficit are the expenses.

Then, TIC minus (trade + budget deficit) represents net profit or loss of the dollar itself. Well, what happens when a company has losses - their stock price goes down... and the same thing has happened with the international value of the dollar since early 2002. When there was a consistent net profit between 1997 and 2001, the dollar value rose.

Some may say that what we're doing is way too simple and there's some truth there... but the bottom line is that it does work and does track and has tracked the value of the dollar for almost 15 years.

tic 1

Click same chart showing trade deficit only

tic 2

TIC short, net

TIC short, net

Click below for raw TIC only charts

World UK Japan Oil exporters Russia Net short term flows Net long term flows
Major Holders Caribbean Korea Historical 2002-5 Treasury monthly raw data TIC flows, 12 month MA, classes 1 TIC flows, 12 month MA, classes 2 TIC flows, totals

(update June 2007 - some good points about TIC data from Brad Setser here.)

Trade and budget deficits, trends

Pain misery index

The simple addition of the unemployment rate to the inflation rate, as measured by CPI.

Note that we have used the U-6 unemployment rate, since the more frequently used U-3 rate data series does not go back to 1900. As of 2007, the average U-3 rate since 1948 was 5.6% and the average U-6 rate since 1947 was 9.5%.

Pain misery index, with housing

Pain misery index, with housing, long term
Pain misery index, with consumer confidence, long term

Yield curves & Recessions

Inverted yield curves, depending on their depth and length, are very highly correlated with economic recessions after one year lag.

It's not "different this time". As of early 2006, the probability is about 25% minimum*.

Recession probability Yield spread - 10 year & 3 month Treasuries
(burgundy line in chart below)
   Recession probability, adjusted **
5% +1.21    30-35%
10% +0.76    35-40%
15% +0.46    40-45%
20% +0.22    45-50%
25% +0.02    50-55%
30% -0.17    55-60%
40% -0.50    65-70%
50% -0.82    75-80%
60% -1.13    85-90%
70% -1.46    95-100%
80% -1.85    100%
90% -2.40    100%
Source - "The Yield Curve as a Predictor of U.S. Recessions"

** Note that other research shows the recession risk to be understated by 25-30% or more by the Fed study sourced above, which means a 50%+ recession chance of one in the U.S. within a year as of mid 2006. Note also that as of late 2006 that recession probabilities have increased greatly in the Euro area.


Click here for longer term yield curve chart.



U.S. credit

Changes in the rate of credit creation tend to lead the economy

Credit, short term




Click here for longer term credit chart.

"Total money"

M3 plus credit, recent time

M3 plus credit, longer term chart

A shorter term chart showing M3 plus credit, plus US derivatives, value adjusted by the Philadelphia Banking Index since its peak in 2007, and here for the same chart but without the Banking Index adjustment.

M3 plus credit plus all government debt, short term

m3, credit, debt short

Add in additional monetization


m3, credit, debt short - bkx

M3 plus credit plus all government debt, longer term charts

m3, credit, debt

m3, credit, debt2

Closeup of M3, credit and gov't debt - 1920-1940

Of course there is some minimal amount of double counting when combining M3 with credit and government debt. But the relative relationships between them stays quite similar over the years, and the amounts are trivial especially when viewing them on an annual change rate basis. Also M3, literally by definition, contains no credit measures at all.
Anyone who believes otherwise is welcome to prove it with actual facts and rigorous analysis, not just links to some Austrian theories or blue sky analysis.

Most of world GDP & money creation (US, Euro area, Japan, China)

Short term

Click here for longer term chart.

GDP & money creation & lags

Changes in the rate of money creation are clearly reflected in GDP, after a time lag.

See money and lags for more data.

For the technically inclined, the GDP projection is based on a lagged rate of change of M3 and volatility plus an Excel forecast of bank credit with debt deflation.

Money supply creation and the inflation link

Click here for the same chart, except going back to 1875.
Click here for the same chart, except with credit instead of m2, and since 1900.

Finally, the broadest & fullest picture of money supply we have
which includes M3 & credit & gov't debt, since 1900.

Global Liquidity

Same chart, except with the dollar index instead of gold

Annual percentage rate of change in the combination of a US money measure called the monetary base plus the total change rate of reserves of the main Central Banks of the world.

It basically measures how fast the central banks are adding liquidity by measuring the growth rate of their own reserves at the IMF, then adding the monetary base to overweight the US. Source: IMF & Fed. Note also that the reserves data on which the charts are based do not include all Central Banks. China, for example, does not report data to the IMF.

A note on this global liquidity chart:
We've had a few question the correlation lag between the two lines in both the late '70s and recently. They ask, if the correlation is supposed to be so good then why was there a 2+ year lag in both cases between the peak in liqudity and the peak in gold. Our answer is related to sentiment based on having wrong facts. Both in the late '70s and recently, most people are not aware or do not believe that inflation is running much higher than what their governments say. When they do start to truly believe that inflation is significant, gold and many other commodities will move much higher.

Be very cautious about extrapolating that gold prices are due to fall greatly *and* on the longer term. We recommend that you notice that global liquidity peaked in 1977 and gold didn't peak until 1980, it's still expanding at over 10% even though the current trend is down, and also that there are strong indications that global liquidity is only temporarily dropping (see GDP and money creation above)(written and as of May 2006).

Annual percentage rate of change obtained by adding the GDP growth rate of the G7 countries, adding the same growth rate percentage data from the global liquidity graph above this one, and then subtracting the average of the interst rates of the 10 year Treasury bond and the 10 year Euro bond.

In other words, we're measuring the production rate of goods and services of the majority of the Western world, adding in excess money creation via the measurement of central banking reserves growth, and then subtracting an average interest rate to account for the cost of the money created and used. Source data is from the IMF, the ECB, & the Federal Reserve.

10 year Treasury bond & Nikkei

The correlation between the 10 year T-Bond interest rate and the Nikkei 225 Japanese stock index is unmistakeable since 1990, but less than good before 1990. Click here for chart.

Stocks vs. hard assets, commodities and US housing

(also known as "The One Chart to Rule Them All")

For the last 200+ years, there has been an average 16 year long cycle where investment returns move between paper and hard asset leadership.
Below we show four ratios: the Dow divided by gold, the Dow divided by 3.5 barrels of oil,the same with the Commodity Research Bureau index, and then the U.S. median house price in gold (times .05 to fit on the scale).

Click here for the same chart, but with the left hand scale reversed from the above. We've found that some prefer it that way.
Click here for the same chart, but only covering the period since 1966. Recent details are easier to see on it.
Click here for the same chart, but only covering the period since 2000. Recent details are much easier to see on it.

Click here for the same chart, but covering the full period since 1800 with as much data as we could find, and here for the same picture with a reversed y axis.

Note 1: The CRB data only goes back to 1956 - data before that has been extracted from various sources and merged in as best we could. It may have significant statistical problems and not be as reliable as we would like. Use it at your own risk. Note also that we switched to the CCI (Continuous Commodity Index) version of the CRB index in August 2006 when the original CRB index was changed, so that historical comparisons remain valid.

Note 2: The housing price data on which the chart was based has been extrapolated and adjusted from the long term Shiller Home Price Index.