by Martin Zweig
+42.5
3/16/72
107.50
-37.6
12/13/74
67.07
+55.4
8/13/76
104.25
+11.4
6/16/80
116.09
+46.3
10/13/83
169.88
+75.5
3/20/87
298.17
-12.9
4/15/88
259.77
+9.2
10/21/88
283.66
+2.2
3/20/89
289.92
+17.5
11/15/89
340.54
+0.9
12/18/89
343.69
-1.6
11/18/90
338.19
+32.5
7/12/94
447.95
+45.1
* 3/20/96
$10,000 becomes: $74,614 $35412
Annualized return = +10.6% +5.4%
Buy-and-hold return = +7.8% per year
The next column on the worksheet grades the Fed Indicator. Recall (see pages 52–54) that I gave it four different gradings based on its past performance. Each of those gradings related to a certain number of “points” on the indicator, which I’ll call “indicator points.” Now, we’ll convert those gradings to “model points” in order to construct our Monetary Model. The listing below shows how this is done.
FED INDICATOR GRADINGS
Indicator Points
Rating
Model Points
+2 or more points
= Extremely Bullish
= 4 Model Points
0 or +1 point
= Neutral
= 2 Model Points
-1 or -2 points
= Moderately Bearish
= 1 Model Point
-3 or fewer points
= Extremely Bearish
= 0 Model Points
Now, refer again to the worksheet in table 9. The Fed Indicator ended 1979 in its “extremely bearish” mode, so it was graded 0. It had -3 indicator points then {not shown in the table). On May 6, 1980, the Fed cut the discount rate, causing the indicator points to rise to +1, a “neutral” rating. Based on the above scoring, the worksheet shows a jump to 2 model points on May 6, 1980. Two weeks later the Fed lowered reserve requirements. That shot the indicator points up to +4, an “extremely bullish” rating. That’s also worth 4 model points. On November 14,1980, the discount rate was raised, dropping the number of indicator points to +1, a “neutral” rating. So, the worksheet shows the model points easing to 2.
Because of the conversion in scoring for the Monetary Model Worksheet, the Fed Indicator’s gradings may be a bit confusing at first glance, but give it a try and you’ll quickly see it’s really simple. Remember this: Changes in reserve requirements and the discount rate (as explained in pages 50–52) give rise to what I call indicator points. The number of indicator points determines ratings ranging from “extremely bullish”e down through “neutral” and “moderately bearish” to “extremely bearish.” There is no “moderately bullish” zone simply because there was no clear pattern of moderately bullish market behavior consistent with any score on the Fed Indicator.
Finally, the indicator points must be converted to model points as shown above in order to develop our overall Monetary Model.
Gradings on the Installment Debt Indicator are much easier. When the Installment Debt Indicator gives a buy signal (see page 66), give it 2 model points. When it gives a sell signal, grade it zero model points. It’s the same process as that done with the Prime Rate Indicator. Note that if the Installment Debt Indicator had a “neutral” rating, it would be given one model point. Another version of this indicator, which I keep myself, has such a range … but in this book I’m striving for simplicity to the fullest extent possible, so I’ve ignored a “neutral” rating. It wouldn’t add much value anyhow.
You ought to keep a worksheet similar to that in table 9. Indeed, you can merely update the worksheet you see there. It’s easy. The prime rate doesn’t change that often, and the rating changes even less often. The Fed doesn’t change reserve requirements or the discount rate very much either. Installment debt figures come out once a month. You have to be exceptionally lazy not to update your worksheet. It’s worth it, and the task ought not to take more than a few minutes a month.
In my Monetary Model Worksheet (table 9), I have included, for illustrative purposes, various market averages (ZUPI, S&P 500, and the Dow). If, when model changes occur, you would like to post and keep track of one or more of these averages, fine. But you don’t have to in order to determine the buy and sell signals. The Monetary Model is merely the addition of all model points. The maximum score is 8, the minimum is zero. There is no way to get a 7 because the Fed Indicator can never hit 3, it runs only 0, 1, 2, and 4. The other two indicators get either a zero or a 2.
TABLE 9
MONETARY MODEL WORKSHEET
Date
ZUPI
S&P 500
Dow
Prime Rate
Fed
Installment Debt
Monetary Model
12/31/79
112.33
107.94
839
2
0
0
2
2/19/80
117.09
114.60
876
0*
0
0
0
5/1/80
103.73
105.46
809
2*
0
0
2
5/6/80
105.52
106.25
816
2
2*
0
4
5/22/80
110.68
109.01
843
2
4*
0
6(BUY)
6/16/80
117.46
116.09
878
2
4
2*
8
8/26/80
132.08
124.84
953
0*
4
2
6
11/14/80
138.18
137.15
986
0
2*
2
4
12/4/80
136.90
136.48
970
0
1*
2
3
12/22/80
132.87
135.78
959
2*
1
2
5
4/24/81
148.46
135.14
1020
0*
1
2
3
6/16/81
149.84
132.15
1003
2*
1
2
5
6/22/81
149.02
131.95
994
0*
1
2
3
9/9/81
128.22
118.40
854
0
2*
2
4
9/21/81
127.04
117.24
847
2*
4*
2
8
2/1/82
128.68
117.78
852
0*
4
2
6
3/8/82
121.07
107.34
795
<
br /> 2*
4
2
8
3/16/82
120.96
109.28
798
0*
4
2
6
6/4/82
125.50
110.09
805
0
2*
2
4
7/19/82
124.03
110.73
826
0
4*
2
6
7/26/82
124.39
110.36
825
2*
4
2
8
5/19/83
203.54
161.99
1191
2
2*
2
6
8/10/83
200.53
161.54
1176
0*
2
2
4
10/13/83
209.04
169.88
1261
0
2
0*
2(SELL)
4/6/84
189.74
155.48
1132
0
1*
0
1
10/6/84
192.21
162.13
1178
0
2*
0
2
10/15/84
194.61
165.77
1203
2*
2
0
4
11/21/84
193.94
164.52
1202
2
4*
0
6(BUY)
1/17/85
205.40
170.73
1229
2
2*
0
4
4/18/86
288.02
242.38
1840
2
4*
0
6
1/9/87
288.78
258.73
2006
2
2*
0
4
3/20/87
322.33
298.17
2334
2
2
2*
6
5/1/87
304.87
288.03
2280
0*
2
2
4
9/4/87
325.23
316.70
2561
0
1*
2
3
11/5/87
231.53
254.48
1985
2*
1
2
5
3/4/88
255.80
67.30
2058
2
2*
2
6
4/15/88
252.98
259.77
2014
2
2
0*
4
5/11/88
249.35
253.31
1966
0*
2
0
2(SELL)
8/9/88
260,07
266.49
2079
0
1*
0
1
10/21/88
264.07
283.66
2184
0
1
2*
3
2/17/89
276.12
299.63
2347
0
2*
2
4
2/24/89
270.45
287.13
2245
0
1*
2
3
3/20/89
271.78
289.92
2262
0
1
0*
1
7/14/89
298.17
331.84
2255
2*
1
0
3
8/24/89
307.51
351.52
2735
2
2*
0
4
11/15/89
288.22
340.54
2633
2
2
2*
6(BUY)
12/19/89
281.07
342.46
2696
2
2
0*
4
1/18/90
283.08
338.19
2666
2
2
2*
6
12/4/90
221.14
326.35
2580
2
4*
2
8
6/22/92
302.75
403.40
3281
2
2*
2
6
7/2/92
307.83
411.77
3330
2
4*
2
8
8/18/92
312.05
421.34
3329
2
2*
2
6
4/19/94
377.33
442.54
3620
0
2
2
4
5/17/94
376.97
449.37
3721
0
1*
2
3
7/12/94
376.70
447.95
3703
0
1
0
1(SELL)
2/1/95
470.40
373.55
3848
0
0
0
0
2/15/95
484.54
384.95
3986
0
1
0
1
7/6/95
424.92
553.99
4664
0
4*
0
4
12/20/95
448.50
605.94
5059
2
4
0
6(BUY)
BUY AND SELL SIGNALS
The Monetary Model is merely the addition of all model points. The maximum score is 8, the minimum is Zero. There is no way to get a 7 because the Fed Indicator can never hit 3, it runs only 0, 1, 2, and 4. The other two indicators get either a Zero or a 2.
You can use the Monetary Model any way you wish to augment other market-timing tools. But for the long-term investor I have devised simple but consistent rules to determine buy and sell signals for the stock market. When the Monetary Model rises to 6 points, it trips a buy signal. That buy remains in effect until the model falls to 2 points, which then flashes a sell signal. The sell then remains effective until the Monetary Model increases back to 6 again, which would trigger a buy signal. To repeat, a buy signal requires 6 points; a sell signal requires 2 points. That’s it.
Table 10 shows all the buy and sell signals since 1954, tracked a
gainst the Zweig Unweighted Price Index. There have been only eleven buys and ten sells since 1954. Each of the eleven buy signals produced profits, including five with gains of better than 50% each. Had you invested $10,000 only in the 340 months that the Monetary Model was bullish (on a buy signal), it would have grown to $392,256, an annualized gain of 15%. That doesn’t include interest you could have earned on money market instruments (such as T-bills or certificates of deposit) during the bearish periods when the model was on a sell signal.
TABLE 10
MONETARY MODEL VS. ZWEIG UNWEIGHTED PRICE INDEX: 1954 to 1996
BUY SIGNALS SELL SIGNALS
Date
ZUPI
% Change
No. of Months
Date
ZUPI
% Change
No. of Months
3/17/54
3173
+53.4
18
9/9/55
51.75
-12.7
26
11/15/57
45.18
+60.7
22
9/11/59
72.61
+1.6
11
8/23/60
73.74
+70.1
66
3/10/66
125.43
-3.3
11
1/26/67
121.28
+20.8
15
4/19/68
146.45
+8.1
4
8/30/68
158.29
+9.4
4
12/31/68
173.15
-41.2
21
9/21/70
101.86
+18.8
21
6/26/72
121.02
-57.7
29
11/28/74
51.21
+78.8
30
5/31/77
91/57
+20.9
36
5/6/80
110.68
+98.1
41
10/31/83
209.04
-7.2
13
11/21/84
193.94
+28.6
42
5/11/88
249.35
+15.6
18
11/15/89
288.22
+30.7
56
7/12/94
376.70
+19.1
17
12/20/95
448.49
+7.2
3
3/20/96 *
$10,000 becomes: $392,256 318 mo. $3636 186 mo.
Annualized return = +15% -6.2%
Buy-and-hold return = +6.5% per year
Assuming an average interest rate of 6% over the period (less in the early years, considerably more since the mid-1960s), you would have earned a total of 161.8% in the 188 months you were out of the stock market. When that sum is compounded onto the stock market return, the original $10,000 investment becomes $1,026,836 in 42 years. That is equal to an annualized return of 11.7%. It still does not include dividends earned while in the stock market. By contrast, buy-and-hold on the average New York stock (as per my Zweig Unweighted Price Index) would have returned only 6.5% a year, again ignoring dividends. Buy-and-hold would have turned $10,000 into only $142,636, nowhere near the $1,026,836 produced by the Monetary Model.
