The Contrarian's View,
Vol. I, #4, November 17, 1986
- PROGRAMMED FOR PROFITS -
One of the things I've noticed about the stock market over the past two
years is the fairly close correlation between stock dividend yields and
short-term interest rates. Of course, stock yields have always followed
the trend of interest rates to some extent, but the "tracking" over the
past two years has been much closer than in previous years... the yield
on the Standard and Poor's 500 stocks consistently being about two-thirds
the current Treasury-bill yield, within a range of 55% to 75% of T-bill
yields.
One of the things everybody has noticed about the stock market in the
past two years is the introduction of "program trading", where computer-
directed buying and selling can change stock prices by 4% or 5% (80 or
100 points on the Dow Jones Industrials) in a single day.
I wondered if there might be a connection between the two. And, after
studying the situation for awhile, I believe there is.... that "program
trading" actually determines the level of stock prices, although within
a wide band where violent fluctuations can occur.
To understand why this is so, you need to see how the computers make
money for their masters. The machines are programmed to profit from
the difference between the current level of stock prices as represented
by, say, the Standard and Poor's index for 500 stocks, and the S&P 500
futures index on the commodities exchange. For example, if the S&P were
at 253, and the future for "delivery" in three months were 256, the
computer would:
a) Use 10% of the available funds to sell the future at 256;
b) Use 90% of the money to buy the stocks which make up the
S&P 500 at 253, in a dollar amount equal to the futures sold.
Over the next three months, the three-point premium of the futures in
relation to the stocks would dwindle away as the futures expiration day
approached. On expiration day, assume that the S&P 500 and the futures
are both at 232; the computer sells the stocks and closes out the
futures position. The computer has thus
* lost on the stocks....................... 21 points
* gained on the futures.................... 24 points
----------
NET GAIN is: 3 points....
....or about 1.2% on the entire transaction (before expenses). If the
computer can do this only four times per year, the annual yield will be:
Yield on S&P 500 stocks.................... 3.28%
Yield on futures premium capture........... 4.80%
----------
TOTAL YIELD: 8.08%
or about 3% more than the yield on T-bills, for what is essentially a
riskless transaction. Since the major expense is the care and feeding of
the computer system, which is minimal when compared with the profits that
can be made, you can see why "program trading" has become so popular.
Investor sentiment.... whether the market will rise or fall over the
near term.... is most quickly (whether rightly or wrongly) reflected in
the futures index. If investors suddenly turn bearish, the index can, and
will, quickly move to a DISCOUNT to the current value of the S&P 500.
Naturally, the computers are programmed to take advantage of this! After
all, why wait until expiration day to close out the position, when it
can be closed out immediately at an even greater profit? The funds can
then be put into interest-bearing T-bills to await the next profitable
arbitrage opportunity.
And this is precisely what happened in mid-September, when the market
nosedived, with the Dow Jones Industrials average shedding 120 points
in two days. The futures indexes moved to a discount (probably because
portfolio managers were "insuring" their holdings against a decline
by selling the futures), and the arbitrageurs' computers sold stocks
and bought futures to close out their hedges. Note that the computers
don't care at what prices the stocks are sold and the futures are bought;
as long as the futures index continues to trade at a discount to the
index itself, the hedge can be closed out at a substantial profit.
Even though the volatility of this computer-directed trading can be
frightening, you can see that it does place nearly absolute limits on
the extremes the market averages can reach. It's almost inconceivable
the yield on the S&P 500 would ever exceed the yield on T-bills, for
the arbitrageurs would quickly move to profit from the higher yield at
less risk, and competition would force the S&P yield lower. (With T-bills
at 5.25%, that's about DJI 1200 for an "absolute" bottom.) Similarly, it's
highly unlikely that stocks would rise so high that the S&P yield, plus the
yield from capturing the futures premium, would be less than the T-bill
rate.
The exact formulas used in arbitrage calculations are deep, dark secrets
locked up in the computer programs of the large trading firms. It is
generally known that they are based on variations of a mathematical
formula known as the "Black-Sholes Method", and also include assumptions
about the future direction of interest rates and the rate of increase of
stock dividends. However, we don't need to know how the computers have
been programmed to profit from the results. We need merely observe that
stocks are now being traded as mere commodities, and that this forces
them to be priced where the yield fluctuates about a level that's around
two-thirds the current Treasury bill yield.
From observation, I've constructed the following table for likely mini-
mum, average and maximum stock prices for different T-bill yields:
T-Bill DJI DJI DJI
Yield Minimum Average Maximum
======== ========= ========= =========
4.5 % 1900 2185 2466
4.75 1800 2070 2335
5. 1710 1960 2220
5.25 1625 1870 2115
5.5 1560 1790 2015
5.75 1490 1710 1930
6. 1425 1635 1850
6.25 1365 1575 1770
6.5 1315 1510 1705
7. 1220 1405 1560
If interest rates remain stable, the market will fluctuate around a
central point - with the fluctuations determined by varying levels of
bullishness and bearishness - but will be essentially trendless, and
extremely frustrating to market observers searching for a trend.
If the Federal Reserve should aggressively drive interest rates down-
ward again.... say to 4.5%.... then load up on stocks, because arbitrage
will eventually drive the market to DJI 2400.
If the Federal Reserve doesn't change its policy, but loan demand slackens
because business conditions are deteriorating, then interest rates will
continue to decline somewhat, and we'll see a most curious phenomenon:
Stock prices will be held artificially high (in relation to the earnings
of the underlying businesses) BECAUSE of the arbitrage. Only if business
conditions get so bad that the stock dividends themselves are cut will
the prices of the shares finally tumble. Of course, this scenario applies
only to the stocks which make up the popular averages; prices of the
remaining stocks will decline in tandem with the deteriorating business
outlook. (This is the stage we seem to be entering right now!)
What happens if T-Bill rates don't remain stable or decline, but actually
start rising? If the rise is gradual, the stock market will probably
decline in a more-or-less orderly fashion, to accomodate itself to the
newly-defined trading range.
One of the disadvantages, though.... in fact, the BIG disadvantage....
of "program trading" is that it doesn't leave much room for error.
Suppose, just suppose, that interest rates don't gradually rise, but
instead there is some sort of crisis. Nothing major, mind you.... just
a small crisis like, say, the failure of Bank of America. Of course, the
Federal Reserve would quickly come to the rescue, as it did with Conti-
nental Illinois in 1984, and the damage would be contained; but in the
scramble for funds, short-term rates would rise rapidly.
Let's assume the mini-crisis raises rates about three-fourths of a
percent from, say, 5.25 to 6%. Suddenly stocks, which have been trading
around the 1870 average, face a DJI 1425 minimum instead of the previous
1625.... 200 points lower!
Portfolio managers, being generally intelligent souls and forseeing an
imminent decline in the prices of stocks, start selling index futures
like crazy to "insure" their holdings against a decline. The futures
quickly move to a discount.... a DEEP discount.... to the actual stock
prices, and the computers in turn go crazy, dumping stocks and buying
the futures the portfolio managers are selling, to mop up the fat profits
from their arbitraged holdings. The bearishness and pessimism become so
pervasive that stocks undershoot the "likely" minimum of DJI 1425, and
head toward the "absolute" minimum of DJI 1060. The worst day sees the
DJI decline more than 200 points; it's off 350 points for the week, 550
points (30%) in three weeks.
The selling panic drives down stock prices well below reasonable levels
(as determined by the earnings and asset values of the underlying
companies). But the computers aren't programmed to be reasonable; they're
merely programmed to make money. The public, however, is quite reasonably
horrified by the stock market crash; responding to the public's hue and
cry, the regulators quite reasonably blame the crash on "program trading",
and act by suspending trading in stock index futures.
Now the trading houses are really in a fix, because they can't close out
the profitable (futures) side of their hedges. What was formerly a risk-
free investment is now fraught with peril, and the arbitrageurs dump
their remaining stocks before they suffer any more losses.
When the dust finally settles, the stock market has declined more than
40%, the public is in a deep funk over its instant loss of wealth, and
a recession is under way.
Could all this actally happen? Well, as I said, I don't see much of a
margin for error.... but I'll let you decide for yourself.
- STOCK MARKET OUTLOOK -
Last month I said I wasn't expecting any major short-term trend in the
stock market, and this certainly has been the case for the past month.
Has anybody died of boredom yet? The two-tier market continues, as the
stocks of the large companies are held aloft by interest-rate arbitrage,
and the prices of the smaller companies' stocks continue to be weak.
Many observers are expecting a December selloff, as investors sell and
repurchase to take advantage of the soon-to-disappear favorable tax rate
for long-term capital gains. Frankly, I think this is being overdone,
since most people are reluctant to sell at all and pay taxes when they
don't have to. However, you can be sure that investors who have losses
are deferring them to 1987, when they can be deducted (up to $3000 worth)
at the 28% tax rate. So I am looking for no great change in stock prices
in December (subject to the usual violent downdrafts and updrafts caused
by arbitrage, of course). But look for a sharp sellof in the market in
January, probably followed by another rally before the bear market begins
in earnest sometime in late winter or early spring.
- WORTH INVESTIGATING -
I almost hate to suggest a look at this stock, because it's done so terribly
over the past several years. TELEFONOS DE MEXICO, S.A. de C.V. (over-the-
counter, TFONY, 3/32) is the Mexican telephone company and a solid, blue-
chip stock. Unfortunately, Mexico is not a blue-chip country. The shares
pay a 12-peso dividend (about 1.428 cents), from which the hard-pressed
Mexican government extracts .7854 cents, leaving the hapless American
shareholder with .6426 cents, or about a 6.9% yield.
To compensate somewhat for the rampant Mexican inflation, Telefonos also
issues a 25% stock dividend annually (in four quarterly installments),
and maintains the 3-peso quarterly dividend on all shares outstanding.
Too bad for the shareowner that Mexican inflation exceeds 60% per year,
so the value of the cash dividend is slowly dwindling away. This has
depressed the stock so much that it currently sells for an unheard-of
1.5 times earnings. This is an excellent example of how hyperinflation
can severely depress security prices.... a superb object lesson for
those who favor liquidating debt via the printing press.
A purchase of this stock is really a gamble that Mexico will bring its
inflation under control, that its economy will recover from the shock
of declining oil prices, and that Mexican stock prices will return to
more reasonable levels. Anybody care to place a bet?
- PORTFOLIO REVIEW -
A. "Hedger's Delight" - model portfolio, includes commissions:
14Nov86
Shrs Description Bought Sold On Sold At Cost Is Value
---- ------------------------------ ------- ------- ------- ------- -------
100 Bally Manuf wt$40/4Jan88 SHORT 30Oct85 281.23 -225.00
1 Bally Manuf 15Sep98cvbd 60.00 2Oct85 795.00 897.50
20 Computer Consoles SHORT 12Jun86 186.74 -170.00
1 Computer Con 15Feb98cvbd 77.50 24Sep85 685.00 780.00
1 Eastern Air L 1Oct93cvbd 47.50 11Sep85 625.00 550.00
100 Golden Nugget SHORT 28Feb86 1390.67 -1037.50
100 Golden Nugget wt$18/1Jul88 28Feb86 357.50 150.00
50 Integrated Resources SHORT 16May85 913.10 -1193.75
15 Integrated Resources SHORT 11Jun85 296.50 -358.13
45 Integrated Resources cvpf 4.25 17Apr85 1560.51 1800.00
25 Integrated Resources cvpf 4.25 15May85 847.50 1000.00
30 Integrated Resources cvpf 4.25 11Jun85 1031.13 1200.00
200 Keystone Camera Products SHORT 12Jun86 1023.86 -1100.00
300 Keystone Camra wt$8.25/20Mar90 11Jun86 330.00 262.50
200 McLean Industries SHORT 5Dec85 1777.90 -575.00
200 McLean Industrs wt9.45/15Jul90 5Dec85 510.00 137.50
165 Pier 1 Inc. SHORT 12Nov85 1663.24 -2908.13
82 Pier 1 Inc. SHORT 23Jan86 931.95 -1445.25
100 Pier 1 wt$22/15Jul88 12Nov85 585.00 2437.50
12 Ridgewood Properties (A) SHORT 12Nov85 252.00 -285.00
100 Safeguard Scientifics SHORT 22Nov85 1010.47 -1575.00
20 Safeguard Scientifics SHORT 6Dec85 222.74 -315.00
100 Safeguard Scient wt$12/30Jun87 18Sep85 220.00 425.00
100 Spendthrift Farms SHORT 9Dec85 341.83 -162.50
100 Spendthrift Farms SHORT 10Dec85 341.83 -162.50
500 Spendthrift Farms wt$9/15Mar89 6Dec85 240.63 93.75
200 Wickes wt$4.43/26Jan92 SHORT 28Aug85 427.48 -550.00
50 Wickes wt$4.43/26Jan92 SHORT 6Feb86 104.06 -137.50
280 Wickes Companies (B) 28Aug85 1226.13 1260.00
CASH 53.75 53.75
(A) Pier 1 distribution; (B) 7:1 conversion ------- ------- --------
of $2.50 preferred. 11159.92 9067.15 -1152.76
SUMMARY - "Hedger's Delight":
Original cost: $ 9,398.02
Present value: $10,007.16
Increase (decrease): $ 609.14 (6.48%)
COMMENT on "Hedger's Delight": There is virtually no change since last
month for this portfolio, which is designed to counterbalance excessively
high or low stock prices. I have specified no trades, and the value is
only slightly greater, due in part to bond interest (Bally, $30; Eastern
Air Lines, $23.75) received.
We do have something unusual coming up, though. One normally expects a
portfolio to yield dividends and interest. Would you believe that "Hedger's
Delight" looks like it's going to produce an airfare discount coupon?
Eastern Air Lines is bribing.... excuse me, enticing.... its bondholders
to vote for an easing of indenture restrictions (as part of its acquisition
by Texas Air) by issuing a $125 airfare ticket voucher for every bond whose
owner votes for the easing at a November 25 special meeting. (Bondholders
who vote against get nothing.) The voucher gets you half price on tickets
(including discounted tickets) purchased before December 1, 1987. I asked
my stockbroker how a ticket voucher gets credited to one's brokerage
account; she was at a loss for words.
B. "Future Income" - includes commissions:
14Nov86
Shrs Description Bought Sold On Sold At Cost Is Value
---- ------------------------------ ------- ------- ------- ------- -------
1 Apache Pet 15Jul2001cvbd 90.00 26Sep86 914.06 840.00
20 McDermott Inter'l wt$25/1Apr90 31Mar86 41.25 62.50
100 Mesa Petroleum 6Feb86 275.00 362.50
50 MSA Realty wt$9/1Apr89 7Feb85 89.38 106.25
50 Public Service NH wt$5/15Oct91 22Jan85 96.25 268.75
50 Varity wt$3.60/31May91 31Mar86 24.07 21.88
------- ------- --------
1440.01 1661.88
SUMMARY - "Future Income":
Original cost: $ 1,467.20
Present value: $ 1,661.88
Increase (decrease): $ 194.68 (13.27%)
Yield: $ 90.00 (6.13%)
COMMENT on "Future Income": A 50% dividend cut on the Apache Petroleum units
has also depressed the price of the Apache convertible bond, and lowered the
return on this portfolio from last month.
C. "Crapshooter's Folly" - includes commissions:
14Nov86
Shrs Description Bought Sold On Sold At Cost Is Value
---- ------------------------------ ------- ------- ------- ------- -------
10 BancTexas cvpf 1.46 (div susp) 21Apr86 44.00 23.75
50 Campbell Resrc wt$4.40/31Dec88 11Sep85 6.00 6.25
50 Damson Energy Partners Class A 8Jul86 116.88 68.75
100 Damson Energy Partners Class B 17Jul86 206.25 137.50
100 Energy Developmt wt$20/31Mar87 31Mar86 13.75 3.13
100 Entex Energy Development 1.00 25Jul86 477.50 387.50
100 Entex Energy Development 1.00 30Jul86 371.25 387.50
100 USX wt$42/15Sep88 6Oct86 68.75 37.50
------- ------- --------
1304.38 1051.88
SUMMARY - "Crapshooter's Folly":
Original cost: $ 1,304.38
Present value: $ 1,051.88
Increase (decrease): $ -252.50 (-19.36%)
COMMENT on "Crapshooter's Folly": The continuing depression in the oil and
gas industry continues to depress the value of this portfolio, which is
heavily energy-weighted. Now you know why I gave this portfolio the title
it has.
D. IRA 1 - real portfolio: Twentieth Century
14Nov86
Shares Description Bought Sold On Sold At Cost Is Value
-------- ------------------------- ------- ------- ------- ------- -------
2.7410 VISTA 8Mar84 11.32 18.66
5.8240 VISTA 21Mar84 25.10 39.66
1.3680 VISTA 31Aug84 6.50 9.32
.0670 VISTA 12Jan85 .29D .46
1.6630 Cash Reserves 12Jun86 166.28 166.28
.0050 Cash Reserves 30Jun86 .50I .50
.0080 Cash Reserves 31Jul86 .80I .80
.0080 Cash Reserves 31Aug86 .80I .80
.0070 Cash Reserves 30Sep86 .70I .70
E: IRA 2 - real portfolio: Bull & Bear
905.8000 Dollar Reserves 16Jun86 900.80 900.80
.9000 Dollar Reserves 30Jun86 .90I .90
2.4000 Dollar Reserves 15Jul86 2.40 2.40
4.5600 Dollar Reserves 31Jul86 4.56I 4.56
4.0800 Dollar Reserves 29Aug86 4.08I 4.08
4.1700 Dollar Reserves 30Sep86 4.17I 4.17
3.9700 Dollar Reserves 31Oct86 3.97I 3.97
--------- --------
$ 1135.17 1158.06
SUMMARY - IRA 1 and 2:
Original cost: $ 952.62
Present value: $ 1,158.06
Increase (decrease): $ 205.44 (21.57%)
COMMENT on IRAs: "Timer's Trend" signalled a BUY on Thursday, October 23.
The formula called for an investment ratio of 10% stocks, 90% cash; the
actual IRA portfolio ratio on that date was 6% stocks, 94% cash. This was
not a great enough discrepancy to take any action, especially when you con-
sider that the market averages are being held aloft by interest-rate arbi-
trage. The fudgefactor has risen from 10% to 15%, primarily because the
current level of interest rates does not adequately compensate for the
underlying rate of inflation.
=========================== TIMER'S TREND ==============================
Wed 17 Sep 86 . #I . | 1769.40 | . - *
Thu 18 Sep 86 . #I . | 1774.18 | .- *
Fri 19 Sep 86 . #I . | 1762.65 | - *
Mon 22 Sep 86 . I .# | 1793.45 |-. *
Tue 23 Sep 86 . I .# | 1797.81 |+. *
Wed 24 Sep 86 . I # | 1803.29 |+. *
Thu 25 Sep 86 # I . | 1768.56 |+. *
Fri 26 Sep 86 . I # | 1769.69 | + *
Mon 29 Sep 86 #. I . | 1755.20 + . *
Tue 30 Sep 86 . I #. | 1767.58 + . *
Wed 1 Oct 86 . I # | 1782.90 + . *
Thu 2 Oct 86 . I# . | 1781.21 |+. *
Fri 3 Oct 86 . I# . | 1774.18 + . *
Mon 6 Oct 86 . I # | 1784.45 | + *
Tue 7 Oct 86 . & . | 1784.85 |+. *
Wed 8 Oct 86 . | # | 1803.85 |+. *
Thu 9 Oct 86 . | #. | 1796.82 |+. *
Fri 10 Oct 86 . # . | 1793.17 |+. *
Mon 13 Oct 86 . # . | 1798.37 |+. *
Tue 14 Oct 86 . |# . | 1800.20 |+. *
Wed 15 Oct 86 . | . # }| 1831.69 |+. *
Thu 16 Oct 86 . | .# | 1836.19 | + *
Fri 17 Oct 86 . | #. | 1837.04 | + *
Mon 20 Oct 86 .# I . {| 1811.02 | + *
Tue 21 Oct 86 . I# . | 1805.68 | + *
Wed 22 Oct 86 . | # | 1808.35 |+. *
Thu 23 Oct 86 . | . # }| 1834.93 |+. *
Fri 24 Oct 86 . |# . | 1832.26 |+. *
Mon 27 Oct 86 . | # | 1841.82 | + *
Tue 28 Oct 86 . | .# | 1845.47 | .+ *
Wed 29 Oct 86 . | . # | 1851.80 | .+ *
Thu 30 Oct 86 . | . # | 1878.37 |~.~+~~~~~~~~~~~~~~~~~~~~~~~~~~~~*
Fri 31 Oct 86 . | . # | 1877.81 | . + *
Mon 3 Nov 86 . | . # | 1894.26 | . + *
Tue 4 Nov 86 . | .# | 1892.44 | . + *
Wed 5 Nov 86 . | . # | 1899.04 | . + *
Thu 6 Nov 86 . | .# | 1891.59 | . + *
Fri 7 Nov 86 . | # | 1886.53 | . + *
Mon 10 Nov 86 . | # | 1892.29 | . + *
Tue 11 Nov 86 . | .# | 1895.95 | . + *
Wed 12 Nov 86 . | .# | 1893.70 | .+ *
Thu 13 Nov 86 . # . | 1862.20 | + *
Fri 14 Nov 86 . | .# | 1873.59 | .+ *
================================================================================
{, } = "Timer's Trend" (4% and 10% exponential) SELL ({) or BUY (}) signal.
[, ] = 4% exponential change unconfirmed by 10% exponential (not a signal).
NEXT ISSUE - will appear about December 15.
/Nick Chase