How To Use Money Supply Statistics For Market Predictions

|
Includes: DIA, QQQ, SPY
by: Austrolib

Summary

Ultimately, what determines stock prices, and all other prices for that matter, is the money supply.

The most accurate picture of the money supply at any given time is the 1-week M2 average, published by the Fed every week.

In the past, a stagnant 1-week average for more than 20 weeks has immediately preceded stock market crashes, as happened in July 2011 and September 2008.

Currently, the 1-week average has remained stagnant for 20 weeks.

The next Federal Open Market Committee (FOMC) meeting will take place on September 16-17. Minutes will be released 3 weeks later. My preferred way to describe an FOMC meeting is this: A small group of people who can print as much money as they want and buy anything they want with it, coming together and deciding how much money to print and what to spend it on.

Justification for their decisions comes in the form of econometric technobabble backed by macroeconomic statistics compiled by bureaucrats about the state of the job market and GDP growth. My question is, if it is moral for them to print money and buy stuff with it for the sake of the economy, why is it called counterfeiting if anyone else does it?

That aside, what never gets discussed in these meetings, at least I've never seen it discussed, is the actual amount of money in the Federal Reserve System itself. By Federal Reserve System I mean the total amount of dollars that exist in every Federal Reserve member bank, be it in the form of savings, demand deposits, small denomination time deposits, or physical currency. How much is there? By how much is it increasing or decreasing as time goes by? One would think that the money supply would be the main topic of discussion for the institution that is in charge of regulating the money supply.

Why is it important? Let's consider a simple thought experiment. Imagine that the money supply were cut in half overnight. What would happen to stock prices? Someone would hit a sell at market order and whoever was thinking about buying beforehand, now only has half as much money as he did yesterday. The stock will fall, not necessarily by exactly half, but very sharply.

Ultimately, beyond all the catalysts and technicals, even fundamentals you can think of, what drives the stock market in the end is the fundamental of fundamentals, and that is the money supply itself.

The Federal Reserve releases statistics on the money supply every Thursday at 4:30pm EST, but never speaks about these statistics to the media. Among Fed watchers there are several ways to monitor these spreadsheets. First let's review what is actually on them.

Table 1

On the first table (the one above is taken from the September 11, 2014 release), you will see a seasonally adjusted vs. a non-seasonally adjusted monthly average of M1 and M2. M1 is currency plus demand deposits. M2 is M1 plus savings deposits, time deposits, and money market funds. The actual number is the non-seasonally adjusted number. Seasonal adjustment tries to average out bumps by assuming the continuation of seasonal trends in the money supply. It is less reliable for getting a picture of the actual amount of dollars in existence at any given time.

Monthly averaging is a good estimation, but it is not the most precise measurement on the release, because on any given day (any given second really) money supply can fall or rise sharply if someone makes a big withdrawal from or deposit into, the Federal Reserve System.

Table Two

Table two shows both seasonally adjusted and non-seasonally adjusted M1 and M2 on 13-week, 4-week, and 1-week averages. If we are concerned with the amount of money available for bidding up the stock market at any given time, the non-seasonally adjusted M2 1-week average is the most precise snapshot-in-time figure available for seeing how many dollars are currently in the system. While the 4 and 13-week averages can give you a better idea of trends, the 1-week is the best for assessing the current situation as it is.

Tables 3 and on show the components of M1 or M2 that are changing week to week.

The question is, what is the best tool for helping you assess the state of the market and your next trading moves? Is it money supply trends as shown by the 13-week average, or more precise snapshots that you can see in the 1-week? This is a matter of debate among Fed-watching investors and traders.

1-Week versus 13-Week Averages

In my opinion, the 1-week average is a better tool for the simple reason that when a big player like a major commercial wants to spend money on stocks, that bank does not withdraw money from a 4 or 13-week average. The bank spends money on stocks from the money supply as it is in the present moment. Granted, that's not the 1-week average either, but that's the closest we can get to it.

The 1-week average is compiled from the average daily tally of M2 over the course of one week. Here's another thought experiment. Imagine two 13-week stretches, one in which the money supply stays constant at $1B (say we're on a gold standard and there is no mining), the other in which the money supply starts out at $1B, and increases by $1B every week for 12 weeks. Then finally, on week 13 some Lehman or other goes bankrupt, there is a giant bank run and money supply crashes back down to $1B. The two 13-week periods would look something like this:

Supply A (Gold Standard)

Supply B (Fractional Reserve Paper Standard)

Week 1

$1B

Week 1

$1B

Week 2

$1B

Week 2

$2B

Week 3

$1B

Week 3

$3B

Week 4

$1B

Week 4

$4B

Week 5

$1B

Week 5

$5B

Week 6

$1B

Week 6

$6B

Week 7

$1B

Week 7

$7B

Week 8

$1B

Week 8

$8B

Week 9

$1B

Week 9

$9B

Week 10

$1B

Week 10

$10B

Week 11

$1B

Week 11

$11B

Week 12

$1B

Week 12

$12B

Week 13

$1B

Week 13 (Bank Run)

$1B

13-Week Average

$1B

13-Week Average

$6.077B

During the first 13-week stretch where money supply stays constant, you would see a 13-week average of $1B. During the second 13-week stretch, you would see an average of $6B. (1B+2B+3B etc. / 13) Looking at just the two 13-week averages, you would see the first average at $1B, and then the second at $6B, conclude that money supply is growing by 600% per quarter, and buy stocks.

However, what is actually happening is that during the previous 12 weeks people were pumping the new money into stocks, and suddenly on week 13 the money is no longer there. Money supply is not growing at all and stocks are about to fall sharply in response to a crash in the money supply. That is why it is crucial to look at the 1-week average to determine your next move rather than the 13-week average, which is only good for spotting broad trends.

If only the real world were as simple as a thought experiment though. How can we use the same logic to predict market movements using the 1-week average? Basically, if the M2 1-week shrinks over a long period of time (20 weeks or more), stock prices will usually go down and perhaps even crash. This is what happened in September 2008. Below is a table of 1-week M2 averages in billions of dollars from April 7th until September 29th, the day of the massive VIX spike.

Apr.

7

7788.6

14

7790.3

21

7771

28

7638.4

May

5

7658.4

12

7669.7

19

7671

26

7649.1

June

2

7684.8

9

7708.7

16

7714.9

23

7659.9

30

7651.1

July

7

7734.6

14

7715.5

21

7705.5

28

7665.6

Aug.

4

7726.7

11

7729.1

18

7720.2

25

7679.5

Sep.

1

7696.1

8

7730.8

15

7747.1

22

7803.6

29

7786

As you can see here, after a period of 25 weeks, the money supply ended of where it began the period, at precisely $7.79T. Contrast that with what was happening in the 25-week period beforehand, from October 8, 2007 until March 31, 2008:

Oct.

8

7394.4

15

7391.1

22

7353.4

29

7355.3

Nov.

5

7414.5

12

7411.3

19

7439.8

26

7440.5

Dec.

3

7465

10

7490.7

17

7505.8

24

7502

31

7512

Jan.

7

7496.7

14

7465.4

21

7457.9

28

7433.4

Feb.

4

7513.9

11

7528.1

18

7544.7

25

7556.1

Mar.

3

7617.1

10

7664.1

17

7706.8

24

7700.3

31

7709.2

Money supply was up 4.2% taking the first and last snapshots of the period. Growth paused for a maximum of only 10 weeks during that period, from November 19 until January 28. Otherwise the trend was consistently up.

Something similar happened during the flash crash of July 2011. In the last week of January 2011, for whatever reason, M2 contracted by 1.4%, a very drastic drop, and was only up by 2.3% 21 weeks later. The money supply started growing after that 21-week break, but 21 weeks was a long enough period of time of no growth to set a crash in motion.

July 2011 Bear

Jan.

3

8885.2

10

8880.2

17

8901

24

8778.7

31

8775.6

Feb.

7

8855.7

14

8871.7

21

8871.3

28

8890.8

Mar.

7

8976.8

14

8987.9

21

8983.6

28

8961.9

Apr.

4

9044.5

11

9053.3

18

9089.7

25

9013.1

May

2

8998.9

9

9030.3

16

9055.3

23

9005.5

30

9025.3

June

6

9104.6

13

9107

20

9118.7

27

9132.1

July

4

9264.6

11

9259.5

Bull Market 2012:

If we compare this to a raging bull market like the one in the second half of 2012, we see no long periods of stagnancy in the money supply. The longest period of no growth was only 11 weeks. Otherwise M2 was up 6.3% over a 25-week period, very strong growth.

June

25

9863.2

July

2

9964.7

9

10009.2

16

10003.1

23

9925.9

30

9942.2

Aug.

6

10037.4

13

10059.1

20

10033.1

27

10004.2

Sep.

3

10103.8

10

10151.3

17

10172.4

24

10053.2

Oct.

1

10154.9

8

10199.1

15

10231.7

22

10168.9

29

10181

Nov.

5

10335.3

12

10303.8

19

10302.4

26

10275.3

Dec.

3

10375.6

10

10427.4

17

10491.1

Using these as benchmarks, we can analyze where we are currently. Here is the current situation, taking a look at a 20-week period beginning April 14th.

Current Situation

Apr.

14

11,353.1

21

11,293.4

28

11,170.4

May

5

11,224.4

12

11,224.8

19

11,218.6

26

11,192.2

June

2

11,304.1

9

11,322.3

16

11,340.0

23

11,249.8

30

11,301.3

July

7

11,381.9

14

11,378.0

21

11,349.5

28

11,323.6

Aug.

4

11,422.3

11

11,394.1

18

11,406.0

25

11,349.1

Sept.

1

11,410.8

M2 has grown by only .5% in 20 weeks. It is basically stagnant. If by next month we see a number similar to $11.35T in the 1-week column, we will be in a very similar situation monetarily speaking to where we were on September 28, 2008. The longer the period of zero M2 growth by the 1-week average, the more dangerous it is for the current price level in the market.

Whichever way you slice it, the one-week M2 average has about 4 to 5 weeks to start growing fast and growing consistently. Any drop below the $11.35T mark from October onward will mean serious trouble.

Perhaps this is why George Soros increased his bet against (NYSEARCA:SPY) to $2B last week. In a month or so, those puts could really start to pay off.

Conclusion

The one-week average - not the 13-week average - is the best picture of the current money supply publicly available. If that number grows consistently without stalling for more than 10 weeks, then stocks can rise, but won't necessarily do so. If that number stays stagnant for more than 20 weeks, we enter a danger zone. If it stays stagnant for more than 25 weeks, we are in a September 2008 situation. The next 5 weeks will tell. Stay tuned.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.