,. TABELL'S MARKET LETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS. INC 1609) 987-2300 August 4, 1989 The market continues to confound its skeptics. The major averages went on posting new bull-markeLhighs..last. week,withtheDowcontinuingto approachitsatime,highof. 2722.42, .the comparable benchmark having been exceeded by the Standard and Poor's 500 a couple of weeks ago. As we have been noting in this space, few signs of internal weakness seem to be manifesting themselves, and the path of least resistance continues upward. It may properly be asked why this should be so. The flip rejoinder is more buyers than sellers. While not entirely true (for each buyer there must be a seller), that response is, at least figuratively, correct. The direct cause of rising prices. market technicians are happy to aver, is the supply/demand equation—a fact insufficiently recognized by those required to invent reasons for market direction. It is true, however, that supply and demand Ultimately rise, at least in part, from perceptions and expectations held by market participants and that these perceptions tend to center around the fundamental factors usually cited as cause for the market's behavior. We discussed one such factor in this space last week—the relatively conservative valuation still being placed on corporate profits despite the market's sharp advance. The S III P 500, we noted, currently stands at just over thirteen times trailing-12-month earnings, a figure whiCh, while not in the bargain range, is hardly excessive based on historical standards. We hear analysts loudly bewailing the fact that bargains exist no longer. a cry not unusual two years into a bull market. The S & P pie belies this plea, although the bargains are probably not to be found among the obvious market leaders of the past couple of years. It IS. of course, the common expectation that earnings may flatten andlor decline over the next twelve months. If this is so. the present modest valuation provides a cushion, and it is worth recalling that earnings and pIe ratios tend to move in opposite directions, mitigating the effect of changes in current earnings on stock prices. Indeed, built into current investor perception is the paradox that the prospect of a slowdown in earnings growth is buIlish—such a prospect enabling the Federal Reserve to allay its primary – – —-concern—rega-rding.,nflationand—-to—pTovide—the–monetaTystimuiusnecessarytocushionan-economic— downturn. giving flesh to that current buzzword, the soft landing. There exists evidence that the Fed may have initiated a policy of monetary ease both in its own affirmations and the behavior of interest rates. Treasury Bills. yielding over 9 as recently as March, now afford a return of well under 8. and the fall in long bond yields has exceeded 200 basis points. Plunging interest rates. the perception of monetary ease, and the prospect of only a moderate business slowdown have, in short, provided an atmosphere conducive to rising stock prices. While the major precept for conduct during a bull market is to relax and enjoy it, the money manager is required by his profession to be a worrier, to consider what might go wrong in an apparently optimistic scenario. One cause for concern would be evidence of mounting deflationary pressure. sugJ!f'sting that the Fed's current prescription for monetary ease might not constitute strong enougl medicine to be effective against the prospective slowdown. Interest rates, as we n;tbO above, have fallen, but there appears. so far, to have been relatively little expansion of the money supply, raising the classical fears about the limitations of monetary policy in producing expansion, fears embodied by the axiom that it is impossible to push on a string. Numerous Cassandras have noted the potential instability inherent in the possible effect of a protracted recession on increasingly leveraged corporate balance sheets. There exists. in other words, the possibility that the soft-landing cushion might ,, pulled out from under financial markets. As is often the case, the exact opposite risk also exists, that the required level of monetary ease might reignite inflationary pressures. Those pressures have quietly been building. the l2-month rise in the CPI as of May being the highest since 1982. The expectation of a hyperinflation seems, it must be admitted, to be clearly absent in the marketplace. As our colleague, Ken Tower, noted this week, gold and commodity prices, presumably sensitive to inflation prospects, have been totally unable to get out of their own way. The market's technical strength, therefore, is based on the expectation of achieving a relatively modest correction in the economy, perhaps, in the process, stemming the recent short-term rise in the rate of price change. Any evidence of failure to achieve these goals could provide early warning signs, caIling into question the viability of the current bull market. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials (1208) S & P 500 (1200) Cumulative Index (8/3/89) 2651.18 344.47 4831.80 AWTebh No statement or expression of opinion or any other matter herein contained IS, or IS to be deemed 10 be, directly or Indirectly, an offer or the solICitatIOn of an offer to buy or sell any secUrity referred to or mentioned The matter IS presented merely for the convenience of the subSCriber While we beheve the sources of our InformatIOn to be rehable, we In no way represent or guarantee the accuracy thereof nor of the statements made herein Any action to be taken by Ihe subSCriber should be based on hiS own trwestlgatlon and information Delafield, Harvey, Tabellinc , as a corporation and Its officers or employees, may now have, or may laler lake, posilions or trades In respect to any securliles mentioned In thiS or any future Issue, and such poSItIOn may be dlHerenl from any views now or hereafter expressed In thiS or any other Issue Delaheld Harvey, Tabell Inc, which IS regIstered With the SEC as an Investment advisor, may gIVe adVice 10 ItS Investment adVISOry and olher customers mdependently of any statements made In thiS or In any other Issue Further informatIon on any security mentIOned herein IS available on request

…. TABELL'S MARKET LETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE. INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC 1609) 987-2300 -. — — – – -……. . . .–T1———Aitl1, 1989' —-….-.— — …-.,;;- – The stock-market story that made headlines this week was the Thursday move of the Dow Jones Industrial Average above 2700 and its momentary flirtation with a new, all-time high—actually consummated as we go to press. Since most other major averages have long ago achieved such highs. this action was hardly earthshaking. If one is seeking truly unusual behavior in a market average, the history of the week1s trading in the Dow Jones Transportation Average is worth a glance. That indicator closed. a week ago, at a level of exactly 1250. Monday's close was 1344.06, a 94.06-point, 7.52 advance. On Tuesday, the Average moved further ahead to 1355.80, making the two-day advance 8.46, and Wednesday's trading took it to a close of 1406.29, making the total three-day advance 12.5. (Another modest advance was achieved on Thursday.) All of the above figures, if not records. were, at least. in record-setting territory. Almost all of this. of course, was due to the action of one stock, UAL Corporation. which became the object of a takeover bid or—in today1s charming language—was put into play. The construction of the Dow Jones Averages is such that, to arrive at the Transportation average. one adds up the prices of the 20 stocks therein and divides by 0.703. Thus, each move of one point in a component is sufficient to produce a 1.42-point move in the indicator itself. Thus, the vast bulk of the Transportation index1s rise can be accounted for by the fact that UAL, which closed at 164 1/2 a week ago, attained closing prices of 210 3/4 on Monday, 19 1/4 on Tuesday, and 243 718 on Wednesday. We will return to this point later,. But first a quick look at the action of the Transportation Average in an historical context. The 7.52 move in that index on Monday is, by far, the largest in the history of the Average since 1949, as far back as our computerized data bank goes. The two-day move to Tuesday has been exceeded only twice in those forty years, and the three-day, 12 112 move, only once—at the 1987 bottom. Looking at the Dow Jones Industnals, for which we have history back to 1926, there have been only 14 larger one-day moves, 24 two-day ones, and 14 three-day instances. With the exception of 1987, all of these cases took placepriorJo..J.19AO…Thus.theweek!s….action-,-in-,-the…..transpor..ts—can-be'8aid–tobea,——I–I modern -record. Yet the cause of all this, the 28 UAL rise, was hardly unique. It was, in fact, just another in a series of takeover-related moves. A scan of our data bank reveals that, looking only at NYSE stocks selling over 20, there have been over 100 one-day moves of greater than 25 since 1982. The left-hand column of the table below lists the 20 most recent such moves and the right-hand column shows the 20 largest. a.OSIHQ P!I1IIOUS I'RCElf( – – – – – -STOCK DOll PRlc a.OSE CWGE a.OSIHQ P!I1IIOUS I'RCElf( – — – – -sm DOll PRlc a.OSE CHAHG UHCIlW AtIII 7 1989 210.750 164.500 28.12 BIIEFICIAL COR!' 111.22 1!SO 71.000 -44.!iOO 61,04 SIlUIIB CW JIJ. 27 1989 112.500 87. 28.21 ICAII STAIIDARI IIIC JAIl 27 1988 58.875 Je.oOO 5MJ llARllIH LABS JIlL 17 1989 33.500 25.250 32.17 PROlIUCTS IIS I CHElUCAI. JIJ. 10 1989 Jl.025 25.125 25.87 IIIRCII1IRT BRA( .lO'J!JIII IUAAI milo IIFG I 18 1997 I 9 1988 .6.625 SJ.750 Jo.sOO , 52.,7 SO,J5 Tl/lIHe JJII 7 1989 170.000 126.000 J4.92 CHVCW AtIII 14 1997 J9.25O 26.125 50.24 CITIWIS I SD1ITlIERH CIl.DIAN COI'IPAHY INC. IIURl.IHGTOH RfS01JRCS 'AA Jl 1m .S7S 26.750 34.11 fJJ 1 1989 7G.125 49.125 2.75 fJJ &1989 49.12'5 J9.25O 25.1& EmJtIRYIHS 8!HK TmS JROC1(1jAI SP 25 1997 711.000 52.250 428 JAM 161m 45.000 JO, 48.76 SP 11 1997 58.m 39.250 48.73 TmS ESTl!Jt CIlW PWSEYLID CflCAGO POCIFlC R J RHABISCO JAIl II 1989 HOU 16 1988 OCT 2 1988 OCT 20 1988 5.ooo .J7'S S6.625 n.25O JO.25O Jl. 4J.J7'S Sl.87S .a.76 2.17 JO.Sl 38.26 J!R;\AlDF.TTlHlHEe. . P!llSB1IRl co SAIIIIIIS !SSOClATE5 OCT 18 1988 OCT 27 1!SO lET 4 1988 JUH 27 1!SO 88.250 29.&25 57.000 SO.SOO 60.125 20.250 J9.ooo 3.750 … 78 46.30 46.15 45.J2 !RAFT lHC. lEST PRWJUTS PlllSBUIIl co OCT 18 1988 88.250 6G.125 46.78 OCT 10 1988 26.875 2'J.75O 29.52 OCT 4 1988 S7.000 39.000 46.15 UJCKT STIlR6 lHe 1lJfOX lHC IIIIIIRE IICCOIIIIACII RfS lIAR 22 19811 .750 32.250 4.91 ..ftJM B 1983 4J. 29.815 4.n FE! Z 1988 34.500 lJ.875 H.SO CO SP 13 1988 SI.25O J9.S00 29.75 SPECT1!A-PItl'lICS )flo 26 1997 3.000 2J.&25 43.92 INSllCO CIlW 111AiI lHe 1llllRA1 01110 ItFIl CAU'.'T Co AtIII J 1988 MY 18 1988 MI 9 1988 ' lJ 1988 2.250 &8.875 SJ.75O 45.375 21.250 50.&25 Jj.7S0 31.875 28.24 J;.os 50.Jj 42.l5 RElO1W1 ICAl ….;I 25 1997 IMOO 42.250 43.20 AffUEi DOTA R&AJtH HOV 18 198 JI.5oo 22.000 .3.18 – CILEIWt COIi'RIJT IHe. CERTAIII !lEI PRODUCT fJJ 14 1989 70.12'i 49.125 42.75 FEB 2S 1988 5.125 JI.&25 2.49 , As the table clearly shows. the UAL rise is hardly unusual. There have been four similar cases in the past month and nine in 1989. It is. in addition. hardly the largest. having been significantly topped by all the moves shown in the right-hand column,. The extent to which takeover-related moves have provided the fuel for the current bull market is a proper subject for a future letter. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials 0200) S P 500 (1200) Cumulative Index (8110189) 2718.13 349.69 4897.72 AWTebh No statement or expression 01 opinion or any other matter herein contamed IS, or IS to bedeemed to be, directly or mdlrectly, an offeror the soliCitation of an offer 10 buy or set! any security referred 10 or menlloned The matter IS presented merety for the convenience of the subSCriber While we beheve the sources of our mlormatlon to be rehable, we m no way represent or guarantee Ihe accuracy thereof nor olthe statements made herein Any action to be taken by the subscriber should be basad on hiS own Investigation and Inlormahon Delalleld. Harvey. Tabelllnc, as a corporation and Its officers or employees. may now have, or may later take, POSitionS or trades In respect to any seCUrities mentIOned m thiS or any luture Issue, and such posl\!on may be different from any views now or hereafter expressed m thiS or any other Issue Delafield. Harvey. Tabet! Inc, which IS registered wrth the SEC as an mvestment adVISor, may give adVice to rts Investment adVISOry and other customers Independenlly of any statements made m thiS or In any other Issue Further Information on any secUrity mentioned herein IS available on request

TABELL'S MARKET LETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS. INC (609) 987-2300 – August 18, 1989 As the market has sailed merrily ahead over the past 22 months, a number of observers—including, -admittedly, ourseleslfave1!lp-reSsed – reservations regarding-thepaucity-of-dailynew highs.,–Just – – recently, on August I, almost two years into the advance, NYSE stocks managed to achieve a bull-market record by posting 306 52-week, daily new highs. In comparison, we may recall that, in October, 1982, two months into the advance which began in August of that year, 653 daily new highs were posted, and, during the next five years, numbers in the 400 to 600 range were not at all uncommon. The table below attempts to provide some perspective by Showing some relevant figures for new highs for the past eight bull markets including the current one to date. For each of the markets studied, the starting low, the eventual high, the percentage advance, and the length in trading days is shown. A number of additional figures are then calculated for the first, second, and third 200-day periods of each bull market. These include the highest daily figure during the 200-day period for new highs and the highest figure for those highs expressed as a percentage of issues traded (smoothed by a 10-day moving average). Also shown are the high for the Dow reached during the 200-day period, and the percentage of the ultimate advance completed at that high. . 6/26/62 10/ 7/66 5/'26/70 1!/ 6174 12817B B/12/82 7124/87 10119/87 21 9/66 ——– -1-2-1–3-/-6-8 –1-/-1-1-1-7-3 –9-/2-1–1-7-6 –V-2-7–/B–l 11129/83 ——– –8-1-2-5-1-8-7 –8-/1–0-/8–9 Low 535.76 70101.32 631.16 577 .60 701.12 776.92 1086.57 1738.74 Huh r. Adv NUlliber of !Iays 995.15 86 913 985.21 1051.70 1014.79 1024.05 1287.20 2722.-42 2712.63 32 67 76 38 66 151 56 518 665 5 798 39 780 458 -D—-1—2-0-0 Peak New Hulhs Peak. New Hlh 184 10.07 229 12.72 298 11.65 393 11.12 284 9.26 653 18.28 3\ ; 100( 3 52 1.81 DJIA H1Sh 706.03 909.63 899.10 S81.81 907.74 1232.59 1299.l6 2J58.61 Advance COtlP, -n,—-'0–'–'0–0- 37 69 64 70 59 89 13 43 Pea, Ne. H13hs 121 251 237 451 174 498 229 PeaI-, New HIgh 6.22 11 .19 10.71 101.11 6.75 16.08 6.05 DJIA Huh 787.78 943.08 950.82 1011.02 893.901 1697.71 2463.89 Advance Camp. 0 0101-600 ———– Pea!-, Ne. Hlhs 55 160 83 76 269 9' 54 224 37 74 332 PeaI-, New HUlh X 10.301 8.99 9.58 11.77 DJIA Hl!lh 991.71 973.51 915.10 1955.57 Advance COIIP. 77 81 61 52 To End Peak New HIghs 157 PeaI-, New Hl!lh Z 10.10 272 13.48 103 4.61 195 278 5.013 10.29 306 9.80 258 7.75 306 10.56 A few observations may be adduced. The first is that the 1982-1987 period appears to be exceptional as opposed to the two decades preceding it. During the five bull markets of those decades, the peak figures for new highs as a percent of issues traded—the fairest way of comparing the statistics over time—tended mostly to be in the 9-12 range. By those standards, the 10.56 level, which represents the high for this market so far, does not appear all that anemic. Markets that score most of their advance in their early phases also tend to produce concentrated peaks in new highs. In the present case, the rise has been fairly gradual. ThIS was also the case in 1962-1966, where only 37 of the eventual advance was completed in the first 200 trading days and just over half of the rise occurred in the first 400. Interestingly, never during the course of that ad vance were over 200 daily new highs achieved. Finally. it must be noted that the shape of the previous bear market has a great deal to do WIth the new high-figures for- thesubsequent upswing. As we all know, the two-month bear market of August-October. 1987 was unique in its shortness, and this made It difficult for the present advance to post large numbers of new peaks in the early stages, since the August, 1987 highs were still being used as a standard for comparison. Most of the other advances followed bear markets that lasted around a year and a half, and levels achieved at the top of the preVIOUS upswing were not taken into account when computing new highs. In addition, it must be noted that only since 1984 have new-high figures been computed on a 52-week basis. Prior to that time they included periods ranging from three to fifteen months. Although the effect of this varies. in many cases it has caused reported new high figures in the early stages of past bull markets to be higher than they probably would have been had they been calculated on today's basis. In the light of all this, current figures for new, 52-week highs, while not stellar, are probably not as bad as they have often been made out to be. ANTHONY W. TABELL Dow Jones Industrials (12 00) 2684.18 DELAFIELD, HARVEY, TABELL INC. S & P 500 (1200) 344.50 Cumulative Index (8118189) 4860.10 No statement or expression of 0p'lon or any ether matter herein contained IS, or IS to be deemed io be, directly or Indlrec1ly, an oHer orihe soliCitation of an oHerto buy Of sell any secunty referred to or mentIOned The matter IS presented merely for the convenience of the subscnber While we beheve the sources of our InformaliOn to be rehable, we In no way represent or guarantee the accuracy thereof nor of the statements made herein Any action to be taken by the subSCriber should be based on hiS own Investigation and Information Delafield, Harvey, Tabelllnc, as a corporation and Its officers or employees, may now have, or may later take. POSItiOns or trades In respec1to any secuntles mentioned In thiS or any future Issue, and such posr\ion may be different from any views now or hereafter expressed In thiS or any other Issue Delafield, Harvey, Tabe!llnc , which IS registered With the SEC as an Investment adVisor, may give adVIce to Its Investment adVISOry and other customers Independently of any statements made In thiS or In any other ISSue Further information on any securrty mentioned herein IS available on request

T Lii.\ IBHEIL.IL.' S IilflI Lii.\!PI l e T IL.IETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC 16091 9B7-2300 – -The4)ow;-finally;-joineU-the-otheraveragesltrhewallAtilung.FuslUt g2l5i,re1ff9i8t9Ory7'!ii'iiIt1ie'OOViou's–,….I…,…… question at' the moment is how far the bull market may carry. In this connection, we were intrigued by the sub-headline of an article on the front page of the Sunday New York Times business section. It read, Will the Dow hit 5000 by 1993. Investigation revealed, however, that the article did not get to interesting topics like the future of the Dow until near the end. Instead, it consisted of an interview with three eminent economists, discussing—pardonably—economics. These three gentlemen, A. Gary Shilling, Albert M. Wojnilower, and Edward E. Yardeni, interestingly enough, held widely differing views. Mr. Shilling is forecasting a recession, ultimately producing long-term Treasury-Bond yields of 4 to 5. Mr. Wojnilower says. flThere is not going to be a recession. and My forecast is that long-term interest rates will go up again. Mr. Yardeni forecasts the best of all possible worlds. no recession and slightly lower bond yields. Long-term readers of this letter will be aware that we have some skepticism about the relevance of all of this, having the usual technician's bias in the belief that the market tends to forecast the economy rather than the other way around. There is even some agreement from the economics profession itself on this subject, since the National Bureau of Economic Research includes the S 81 P 500 in its series of leading economic indicators. The gentlemen in the Times interview, however, disagree. Mr. Wojnilower contends, flVery rarely do the securities markets tell you anything useful about the economic outlook. 11, and Mr. Shilling noted that The market didn't predict 1929, or the 1980 recession. We hope we will be forgiven for remaining unconvinced. Although the message was certainly not clear, the market remained in a trading range—with nasty shakeouts in fall, 1978 and fall, 1979—for two years prior to the 1980 recession. Furthermore, correctly forecasting that recession would have been of limited usefulness since, starting from the recession's mid-point, the Dow posted a healthy35, yearolong gain. Thestock market admittedly,diLl1ot.o- I lead the economy in 1929, although the severify of that' break has iong been associated with theeventual depth of the 1929-1932 depression. This, of course, is what made similar market action in 1987 so scary. As the most bullish of the three interviewees, it was, unsurprisingly, Mr. Yardeni who looked for the Dow to be at 3000 by year-end and suggested, We are on our way to a 5000 Dow by 1993. We are not ready, at this stage, to disagree with this forecast, but it is interesting to note, in our capacity as market technicians, that such a forecast could have been arrived at via a naive approach, having nothing to do with economics. Readers are familiar with our compilation of the 24 major stock-market cycles which have taken place since the Dow was first computed in 1896. Measured using monthly average prices. these cycles have a mean length of 46 months, measured from low to low (thus the term tlfour-year cyclell ), with the advancing phase averaging 30 months. The average percentage advance for t!le 24 cycles has been 81 and the mean subsequent decline, 28. Overall, the low for each cycle has tended to be 26 above the previous low, and the highs have averaged out to 31 higher than the prior high. Given these figures, all we need to know is that the average price for December, 1987 was 1910.07, and we can arrive at the following scenario. Thirty months from December, 1987 takes us to June, 1990, by which time the Dow, if it advances 81, would be at 3457. A subsequent 28 decline would bring it to 2489, which figure, were it reached 46 months after December, 1987, would be attained in October, 1991. Why not go on Given the projected benchmark low of 2489 in October, 1991, we can then extrapolate a top at 4505 to occur in April, 1994. Now we are certainly not naive enough to offer the above as a forecast, and we have regularly noted the extreme variability of market patterns. Nonetheless, we do not think a composite of the market's behavior over nearly 100 years constitutes nonuseful information. What is interesting is .. the.plausibility of Mr.Yardeni'sforecast in.thelightof-the-..,.. historical record. What could go wrong We are, of course, not total disbelievers in the usefulness of economics, and what bothers us is Mr. Shilling's projected recession or, more properly, the possibility that it might turn into something even more serious than he projects. Were such to be the case, it is of course, our faith that the market would provide us with some prior warning in the form of deteriorating technical action. It is obvious, though, that such deterioration has not manifested itself to date. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials (1200) S & P 500 (1200) Cumulative Index (8124189) AWTebh 2740.13 351. 73 4901.15 No statement or expression of opinion or any other matter herein contained IS, or IS to be deemed to be, directly or indirectly, an offer orthe SOlICitation of an offer to buy or sell any security referred to or mentioned The matter IS presented merely lor the convenience of the subSCriber While we beheve the sources of our InformaltOn to be reliable, we In no way represent or guarantee the accuracy thereof nor 01 the statements made herem Any acltOn to betaken by the subSCriber should be based on hIS own Inves\JgallOn and information Delafield, Harvey, Tabelllnc, as a corporation and ItS officers or employees, may now have, or may later take, poSlltOns or trades In respect to any securities menlloned In thIS or any future Issue, and such POSition may be drtferentfrom any views now or hereafter expressed m this or any other Issue Delafield, Harvey, Tabelllnc which IS registered With the SEC as an Investment adVisor, may 91ve advice to Its Investment adVISOry and other customers Independently of any statements made In thiS or In any other Issue Further information on any secunty mentIOned herein IS available on request

1Li\ (lUE IL.IL.' S Li\IRlIE1 1L.lE11lEIRl 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC (609) 987 -2300 September 1. 1989 Just about a year ago in this space. we remarked on what seemed to us at the time to be a rather strange phenomenon—the large number of issues that seemed to be windlng up each day unchanged from tnei!, – previous cIose-.-WIlat(\rew- this -toQurattention- was-the-fact that- 0I1mll.ny-days ;-mor,,'thaI1500– – issues were closing unchanged. something that had formerly been a fairly rare occurrence. We decided at that point to go back and look at some past history. Quite obviously, the round number of 500 was an accidental product of the number of issues that at the time tended to trade on a given day—between 1950 and 2000. For purposes of historiCal comparison, 500 could be translated into 25.5 of issues traded. What was remarkable to us about our examination was the dlscovery that. from time to time. trading patterns seemed to emerge where unusually large numbers of issues tended to trade unchanged. these patterns remaining in force for protracted periods. Updating the figures, there have been, through yesterday, 17150 trading days since 1926, and on 2832 of these more than 25.5 of all issues traded were unchanged. Our year-ago letter noted that well over three-quarters of those days had occurred in clusters of 100 days or more. a cluster being defined as a continuous series during which there occurred no two instances more than 25 days apart. The largest of these was 1063 days occurrmg over a SIx-year period between 1939 and 1945, and there was a 342-day string in 1951-1954. The cluster which we identified a year ago has continued since that time. Yesterday was the 200th trading day (the first one was April 27, 1988) on which more than 25.5 of issues traded were unchanged for the day. There are, conversely, long periods when the pattern fails to emerge. The phenomenon of 25.5 of stocks unchanged in a day never occurred between 1958 and 1974. If more stocks are indeed trading unchanged. there is an obvious corollary; there must be fewer advances, fewer declines, or both. TentatIvely, the answer appears to be both. Advancing days over the past couple of years have seemed to feature fewer advancing stocks. but declining days have also produced fewer declines. The following table summarizes some figures. Dow Change -2 or more -2 to -1 -1 to Unch. Unch. to -1 1 to 2 2 or more No. of Cases 1344 6255 7154 1402 482 1 9-.1L6 9-.89 Average of Average of Advancing Stks Declining Stks 13.1 71.3 20.3 61.4 32.3 45.8 45.8 32.2 57.8 23.3 66.4 17.7 L9IUt.L9–1lJ. No. of Average of Cases Advancing Stks -9- 13.9 34 22.8 149 33.4 167 41.9 46 50.7 16 61.7 Average of Declining Stks 68.8 54.3 41.0 32.3 25.5 18.7 — -. The fIrst three columns in the table above cover the entire 64-year period from 1926 to date, whereas the second three columns only cover 1988 and 1989 thus far. We have assIgned each trading day for the two periods to one of six brackets based on the change in the Dow. For each group of trading days, we have shown the average number of advancing stocks expressed 8S a percentage of issues traded and the comparable statistic for declines. Thus, the table shows that, for the 16 days since 1988 on which the Dow advanced 2 or more, the average number of advancing stocks was 61.7 of issues traded. Over 64 years, the advance percentage on such days had averaged 66.4. In the past two years, when the Dow advanced between 1 and 2, just oVer half of all issues traded rose. Over the longer period, 57.8 of stocks changing hands tended to advance when the Dow rose this amount. No year since 1926 has exhibited such a low advancing percentage for this bracket. However, albeit to a somewhat lesser extent, the reverse seems to be true on declining daysw The table quite clearly shows that, during the 1988-1989 period, on days when the Dow declined, there were fewer declining stocks than would be suggested by the 64-year trading pattern since 1926. For the current bull market, it would seem, movement in both directIons has been confined to 8 relatively restricted number of stocks. This can be rationalized in a number of ways, the obvious one being to point to the concentration of upside action in takeover stocks. There is at least one additional possible explanation in the hypothesis that markets today are broader and more liquid than in the past, thus producing a higher degree of short-term price stability. Neither of those explanations, of course, accounts for similar periods in the early 1940's or mid-1950's. It is not clear what implIcations all of this may have for a market forecast. A year ago, we cited evidence to suggest that. over a long period, the implications were bullish. We see nothing in the additional research above to suggest otherwise. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials (1200) S & P 500 (1200) Cumulative Index (8/31/89) AWTebh 2756.08 352.92 4934.04 No statement or e)(presslQn of opInion or any other matter herein contained IS, or IS to bedeemed to be, dlrectty or indirectly, an offer or the soliCitation of an offer to buy or sen any security referred to or menhoned The matter IS presented merety for the convenience of the subSCriber While we beheve the sources of our information to be reliable we In no way represent or guarantee the accuracy thereof nor of the statements made herein Any action to be taken by the subSCriber should be based on hiS own Invesllgatlon and InformatIOn Delafield, Harvey, Tabelltnc , as a corporation and lIS oH!cers or employees, may now have, or may later take, poSitions or trades In respect to any seCUrities mentIOned In thiS or any future Issue and such poSItion may be different from any views now or hereafter expressed In thiS or any other Issue Delafield, Harvey, Tabellinc , which IS registered WIth the SEC as an Investment adVisor, may give adVice to lis Investment adVISOry and other customers Independently of any statements made In thiS or In any other Issue Further information on any security mentioned hereIn IS available on request

.- TABELL-S MARKET LETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC (609) 987-2300 H, September 8, 1989 ….,..,1—–..We-try I intl1isspace,. to av'Qicrthele-mphition.-W.to-turn- ayunctilarbu-rwecannotchangf.f thef8Gt -0.. that we have been writing financial commentary (and reading it) for more than 35 years. Since there is little new under the sun, it is easy to predict the sort of thing one will hear after the market has been going up for quite a while as is the case at present. As surely as death and taxes, there will emerge pronouncements informing us that the market is in need of a correction. We are not sure why this is so. It may be that there remains buried deep in our psyche Borne remnant of the puritan ethic, which, essentially, holds that it is sinful to have fun. It IS almost as if a rising market constituted a manifestation of illness which could be cured only by periodic doses of unpleasant medicine. However, a look at the facts of market history suggests that protracted. uncorrected rises are, first of all. perfectly normal phenomena and. secondly, tend to be signs of market health rather than sickness. In the present case, the period of time during which the market has failed to undergo a needed correction can be dated back to November 16, 1988. Since then, over exactly 200 trading days, the Dow rose 35 to a high reached a week ago without a single correctIon of as much as 5, the biggest interruption being a three-week hiatus involving a 4.44 correction last March. Since that time, the rise has been even steadier. with the Dow moving ahead 22.69, while the largest correction was a 3.63, five-day affair in June. The point that needs to be made is that this behavior is in no way unusual. Every bull market in modern stock-market history has produced a protracted rally not unlike the present one, and in many instances, that rally has greatly exceeded the current advance in both time and extent. The table below shows some figures. Bull Market Start Start Date Date Jun .13 49 J.un1349 Sep 14 53 Sep 14 53 Oct 22 57 Dec 17 57 Jun 26 62 Nov 22 63 Oct 7 66 Oct 7 66 May 26 70 Jul 7 70 Dec 6 74 Oct 1 75 Feb 28 78 Apr 21 80 Aug 12 82 Jan 24 83 Jul 24 84 Dec 7 84 Oct 19 87 Nov 16 88 LON G EST RALLY Days After Percent Length Bull Market Bull Mkt Start Advance In Das End Date End Date 0 41.32 282Jun12…s.0 –.JanL53 0 60.04 326 Jan 3 55' -Apr 6 56 38 59.31 410 Aug 3 59 Dec 13 61 356 32.06 370 May 14 65 Feb 9 66 0 22.21 145 May 8 67 Dec 3 68 29 42.05 205 Apr 28 71 Jan 11 73 206 28.93 140 Apr 21 76 Sep 21 76 542 28.38 107 Sep 22 80 Apr 27 81 114 21.17 100 Jun 16 83 Nov 29 83 96 59.55 343 Apr 21 86 Aug 25 87 274 35.00 200 ' Days Later 705 318 595 187 373 431 106 149 115 341 – Shown in the table is the longest rally for the ten most recently completed bull markets plus figures for the current one to date. For the purposes of this exercise, a rally is defined as an advance without any 5 correction. Six of the past ten bull markets produced rallies of this type lasting longer than the current one has so far, the longest of these running for 410 trading days, from December, 1957 to August, 1959. Five such rallies produced greater percentage advances than this one has shown so far, including rises of approximately 60 in 1953-1955, 1957-1959, and 1984-1986. The other advances were not too different in character than the current case, the smallest one being a five-month, 21 rise in January – June 1983. Through the mid-1960's there appeared to be a tendency for the longest rally of a bull market to be the first one, starting either from the low or from a test within a couple of months of that low. More recently, the tendency seems to have been for the longest rally to begin much later (viz December, 1957 or July, 1970). This one, having started November 16, 1988, over a year after the October 19, 1987 low, is typical. 1 What,is. perhaps, most -interesting is -that..-inno'casehas ,the-endof'along'rally signified the — – bull market's end. The last three columns of the table show the end date of each upward cycle's longest rally, the actual end date of the cycle, and the intervening number of days. The shortest interval between the rally's end and the bull-market peak was five months in April – September, 1976, and there have been cases where the advance continued for a8 much as two and a half years following the long rally's completion. At the current stage, therefore, although a short-term correction would be consistent with past history, there is no particular reason to expect one. Such a correction, moreover, were it to occurJ would, at worst, constitute an extremely early warning of the termination of the present bull market. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials (12 00) 2693.20 S & P 500 (12 00) 346.13 Cumulative Index (9/7/89) 4929.50 No statement Or expression ofoplnlOO or any other matler herem contamed IS, or IS to be deemed to be, directly or Indirectly, an ofter or the sollCltalion 01 an offer to buy or selt any secUrity referred to or mentfOned The matter IS presented merely for the convenience of the subSCriber While we believe the sources of our InformatIon to be reliable, we In no way represent or guarantee the accuracy thereof nor of the statements made herem Any action to be taken by the subSCriber should be based on hiS own Investigation and InformatIOn Delafield, Harvey, labelllnc, as a corporation and tts officers or employees, may now have or may later take, poSitions or trades In respect to any securities mentIOned In thIS or any future Issue, and such posrtlon may be dllferen\ from any views now or hereafter expressed In thiS or any other Issue Delafteld, Harvey. labell Inc , which IS regIstered WIth the SEC as an Investment advisor, may give adVice to ItS Investment adVISOry and other customers Independently of any statements made In thIS or In any other Issue Further Information on any security menboned herein IS avaIlable on re9uest

'1l'IBUE D.D.' S RlIE'1l' D.1E'1l''1l'lElRZ 600 ALEXANDER ROAD. CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC (609) 987-2300 September 15. 1989 – – – – W e .have ..regularly-em8.l'ked,-inthis.s pace .thaLone, ofthefunctionsof the.tech nician.is to – I serve as market historian. In'this capacity, his task is to seek out prior periods of market history which bear some resemblance to the current one and to adduce from subsequent results the elements of a forecast. We have found ourselves discussing history a good deal of late. This is appropriate since it is fairly easy to fit the stock-market scene, 8S it exists in September. 1989, into an historical context. There can be little doubt that October 19. 1987 (or its December. 1987 test) constituted a major bear-market low. We are now approaching the second anniversary of that low. This puts us in an almost two-year-old bull market which has. to date. scored an advance of some 58.2. Now bull markets—and the inevitable bear markets which follow—are not new phenomena. They can be traced back as far as the Holland tUlip market in the 1600's. and we have noted. most recently just three months ago. that there have been 24 measurable bull markets over the past century. Certainly. a study of those 24 previous markets should tell us something about the present case. It does. indeed. do so. but it also suggests some of the limitations of historical analysis. It is quite true that the average length of the bull phase of a major cycle is 30 months and that many bull markets have had lives approaching four years or even longer. Nonetheless. there exists a fair number of recorded cases in which 8 bull market has topped out in 24 months or less. There are, to be exact. ten such instances. although most of them occurred early in the century. We tried to carry historical analysis a bit further last week. when we noted that. since last November. the Dow had advanced for 200 trading days. to September 1. without a 5 correction. The major point we wanted to make was that. following past similar rallies. the first 5 correction had never constituted the end of the bull market. The above, it seems to us, affords an historical framework. We find ourselves in 8 major-cycle upswing which probably has more room—possibly a good deal more room—on the upside. but one where —-the-inevitableoppin grocesscould-begin-a t -any time-' To'determine'whetheroronot-that'process—- has begun, it is necessary to focus more closely on short-term market action. In terms of the Dow. a potential top does exist. with a downside objective of 2560. This top. however. would not be confirmed unless the Dow were to break below the 2630 level. If the downside objective is reached, it would constitute the first 5 correction since last November, an event which should have considerable lead time on a market top. In terms of individual-stock action. which for ten months has been confined almost exclusively to the upside, the market has turned mixed. We track each day using a computer scan, all upside and downside breakouts and reached price objectives for some 5.000 individual stocks. This Wednesday was typical. While the Dow ended that day sharply lower. there were a total of five upside breakouts and two downside objectives attained, or seven bullish events. On the same day seven stocks broke out on the downside and four reached upside objectives. for a total of eleven bearish occurrences. This sort of thing has been continuing for a couple of weeks. For the time being, at least. most deterioration has been confined to the formation of short-term tops in the defensive industry groups that have. by and large. been market leaders over the past year. food stocks being notable examples. At the same time. upside breakouts and ongoing uptrends continue to manifest themselves in cyclical. heavy-industry issues and in energy stocks. This action. interestingly enough. suggests that the market is forecasting a different economic scenario than the one generally pictured in the financial press. That particular world-view is greatly concerned with making distinctions about hard and soft landings and seems confident that the authorities have inflation well under control. In the meantime, the market's shift in leadership suggests that the economy has no intention of landing at all. but intends to fly for a while longer. It also suggests that the expanding inflation figures for the f,rst five months of this year are more indicative of the outlook for the price level than the smaller increases shown for June and July. , – – In summary, the first significant correction in over ten months appears to be, at least, a possibility. Such such a correction would be unlikely to signify the end of the bull market and should be followed. if it occurs. by a move to new highs. Concomitant with this action. upside activity should become less broad as market leadership shifts into new areas. ANTHONY W. TABELL DELAFIELD. HARVEY. TABELL INC. Dow Jones Industrials (12 00) S & P 500 (1200) Cumulative Index (9/14/89) 2666.60 342.52 4884.65 AWTebh No statement or expreSSion of oplmon or any other matler herein contained IS, or IS to be deemed to be, directly or Indirectly, an offerorthe soliCitation of an offerlo buy Of sell anysecurrty referred to or mentioned The matter IS presented merely for the con'lemence of the subSCriber While we belie'le the sources of our information to be reliable, we In no way represent or guarantee the accuracy thereof nor 01 the statements made herein Any action to be taken by the subscnber should be based on hiS own investigation and Information Delafield, Harvey, Tabeil Inc, as a corporabon and ItS offICers or employees, may now have, or may laler lake, poSitions Of trades In respect to any secuntlas mentJoned m thiS or any future Issue, and such poSition may be different from any vIews now or hereafter expressed In thiS or any other Issue Delafield, Har'ley, Tabellinc , which IS registered With the SEC as an Investment adVIsor, may gIVe adVice to Its rnvestment adVISOry and other customers mdependently of any statements made In thIS or In any other Issue Further mformatlOn on any security mentIoned herein IS available on request

T J.ii.\1mIEn.n.'S J.ii.\ R IlllET n.lETTIER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC (609) 987-2300 – September 22, 1989 -f —As .pwl'lDrth'l8sCp,.epres,-iobe buJLeted byh-lr-,1al1e,JI-,!go,a few s.10rllL.ql.o,uds-lu9kilyno,t,….,..,.., of hurricane'strength—are appearing on the stock-market scene. It seems clear that the market – has lost some of the ebullience that had characterized it until early August. It was on August 10, six weeks ago, that the Dow first moved above 2700. There have been subsequent moves to new highs, carrying the Average above its 1987 peak, but they have been somewhat sporadic. Meanwhile, twelve trading days have elapsed since the all-time-record peak achieved on September 1 at 2752.09. This is, interestingly, the longest gap between new highs since mid-April While all this has been going on, a minor breadth divergence has appeared, with no new high being posted by our daily breadth index since August 8 despite a series of new peaks for the Dow. It may be, of course, that we are seeing nothing more than yet another iteration of the seasonal pattern for September. This pattern, which, based on recent history, is the strongest one currently extant, calls for a decline during that month. So far, the market seems to be conformlng. with the bull-market peak. as noted above. having been attained on September's very first day. It may be worthwhile to recapitulate the evidence for this pattern. Since the first computation of the Dow in 1897, there have been 622 up and 476 down months. In other words, some 55 of all months in the past 92 years have been up ones By contrast, of the 91 Septembers during the period, only 36 have produced a rise in the Dow, and 55 have produced lower prices at month's end, the average decline for the month being 1.28. If anything, the tendency has been exaggerated in recent years. Since 1969, there have been only three Septembers which showed rising prices. Set against this background. we have witnessed the relatively rapid formation of distributional tops, first in a few individual stocks and now, potentially at least, in the market averages. The Dow, as noted above, first moved above 2700 in early August. It then pulled back to as low as 2640 on August 15 and, after reaching new highs, has moved back again into the –'2650-2700area.-If–thisformatiorr–is-to4eeadsa'-distributional-top.theownside—target.—-…- would be approximately 2540. This target would not become effective, it must be noted however, until the 2640 level is penetrated. The patterns for all of the averages appear to be remarkably similar. Objectives for the various indices—recent levels are in parenthesis—are as follows Dow Jones Industrials (2682), 2540; Dow Jones Utilities (217), 206; Standard & Poor's Financial (33.00), 31.30; Standard & Poor's Industrials (396), 375; Standard & Poor's 500 (346), 328. All of these targets involve drops of around 7 from recently attained highs and of a bit under 5 from current levels. They would, furthermore, become effective only if the mid-August lows, comparable to 2640 on the Dow, were penetrated. As we noted above, the currently visible storm clouds are hardly of hurricane proportions. At the moment the worst-case scenario calls for nothing more than the first minor correction following a 200-day, uninterrupted, bull-market leg. We noted in our letter last week that such corrections have not, in the past, tended to constitute the end of the bull markets in which they occurred. They have, rather, tended to be the initial elements of protracted topping processes, which have often taken many months (in some cases well over a year), and have, invariably, involved new highs in the averages. There remains, of course, still unexorcised, the ghost of 1987. As we have pointed out in this space, the characteristic feature of August-September during that year was the unprecedentedly rapid build-up of distributional top formations. As tops build up with similar rapidity almost exactly two years later, the spectre of a replay of that grim scenario must loom large in the analyst's thinking. We do not think it will occur. Based on current evidence, we think that the topping-out process for this bull market is likely to be the conventional, lengthy one, involving levels areabove—possibly well above—the -iit-variably-corifounds thosewho 2752 peak already attained on the Dow. The market looking for it rephiyoCrecent historY-Thus, the almost cur-r e nt— – – – – I – – market will, we think, either hold above the mid-August lows and proceed expeditiously to new highs, or—as outlined in the scenario above—break those lows and embark upon a minor correction. That correction should be just deep enough to awaken, in timid souls. the memories of 1987. For the courageous, it will constitute a buying opportunity. Dow Jones Industrials (I2 00) S & P 500 (1200) Cumulative Index (9/21/89) AWTebh 2677.24 345.74 4885.12 ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. No statement or expression of opinIOn or any other matter herein contained IS, or IS to be deemed to be directly or Indirectly, an offer or the solicitation of an offer to buy or sell any secUrity referred 10 or mentioned The matter IS presented merely for the convenience of the subSCriber While we beheve the sources of our information to be reliable, we In no way represent Of guarantee the accuracy thereof nor of the statements made herein Any action 10 be taken by the subscriber should be based on his own Invesllgatlon and Information Delafield, Harvey, Tabell tnc, as a corporation and Its officers or emptoyees, may now have, or may later take, positions or trades In respect to any securities mentioned In thiS or any future Issue, and such poslbon may be different from any views now or hereafter expressed In thiS or any other Issue Delafield, Harvey, Tabell Inc, which IS registered With the SEC as an Investment adVisor, may give adVice to Its Investment adVISOry and other customers Independenlly of any statements made In thiS or In any other Issue Further information on any security mentIOned herein IS available on request

TABELL'S MARKET LETTER 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC 609) 987-2300 September 29, 1989 The Dow Jones Industrials, staging a sharp, mid-week rally, continued to hold above the 2640 -……,… – – level-;–wtlich-represents'-the …-low reached-in- Au gus t–'and-..-w hich-would. constit u te-'-a–downsidereako1J.t . from what now must 'be considered a fairly important potential top Continued ability to hold in this range without breaking 2640 (or 342 for the S & P 500) could only be construed as bullish and would indicate that the present interruption in the advance which began last November would be completed with only a consolidation rather than a correction. Unimpressive breadth and volume, however, continue to suggest the likehhood of a downside breakout eventually taking place. The downside targets for such a decline, were it to occur, are not, at the moment. all that serious—centering around the low 2500's. The problem would be to fit it into a longer-range scenario. As technicians, we are, of course, committed to the belief that market patterns repeat themselves and that bull and bear markets tend to follow similar rules of behavior from cycle to cycle. There are, of course, just enough exceptions to this rule to make life interesting, but until evidence to the contrary accumulates, it is wisest to treat the exceptions as being just that. Up until a few years ago, at least, there existed a reliable rule of thumb concerning major tops and major bottoms. The former generally built up over'a protracted period of time as a given advance slowly ran out of steam. accompanied by deteriorating measures of market momentum. The latter. by contrast, could be identified by obvious selling climaxes, sharp declines and recoveries on increased volume compressed into a short time frame. Past major lows—June 1949, September 1953, October 1957, May – June 1962, October 1966, May 1970, and October – December 1974—all were identifiable by such climaxes. In the past 15 years, bottoms have emerged in a quite different form. February 1978 and August 1982 (also July 1984, if one chooses to treat it as a major cycle bottom) displayed a new set of characteristics. They began with a slow drift to lower levels on decreasing volume and then, out of the blue, put on an upside explosion. The older pattern of a selling climax was conspicuous by its abThseeclimnacctic ebot.tomreturn7edwith avengeance onOc-tober121l7l911'7o' but, follOWing thetest–''-'I of the climactic lows in December, 1987, the market displayed behavior quite different than that shown following earlier climaxes. Between 1949 and 1974, after the low (and its test if there was one) had occurred, there ensued a long rally, generally the strongest one of the bull market. This, as we all recall, was not the case with the current upswing. It spent most of the year 1988 swinging back and forth in a trading range with a moderate upward bias, and not until last November did we witness the characteristic bull market extended leg, a phase which which mayor may not have ended on September 1. Interestingly, another recent market cycle, 1978 – 1981, behaved in more or less the same fashion with a protracted trading range running from February 1978 through Silver Thursday in April 1980, at which late date a long upside leg developed. Thus 1987-89, in this respect at least, behaved in the modern fashion. What about the characteristics of market tops, which are now what we should be concerned about following a two-year advance. As we noted above. we had learned, over the years, to expect tops to form over a lengthy period of time and, during that time, to provide the analyst with a battery of evidence for detecting them. Then, of course, there occurred August – September, 1987, whose salient characteristic, as we have noted many times. was the amazing speed with which tops built up. The only historical parallel for this, was the summer of 1929, and, as far as the break was concerned, we did indeed have a replay of 1929. The aftermath, of course, was quite different—for which blessing we may all be truly thankful. The interesting question, of course, is whether 1987 was an exception to the rule or constituted the first exemplar of a change in the pattern for market tops, much as 1978 seems to have initiated a change in the typical bottom formation pattern. For the time being, at least, we remain inclined to treat it as an exception, and it is our current view that any short-term correction which might -occur over-the near term …would be nothing more than a correction. As we indicated.last.week. we.do not think that the ghost of 1987 has been laid to rest in the minds of investors, and we think that any noticeable decline at this stage, which, as noted above. we consider probable, would bring recollections of 1987 to the surface. For the time being, at least, we remain convinced by the historical evidence which suggests the current bull market stil! has further to go, and, unless the current top broadens, it remains our view that a short-term downswing represents the worst-case scenario at this time. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. Dow Jones Industrials (12 00) S P 500 (12 00) Cumulative Index (9/28/89) AWTebh 2704.21 349.56 4901.15 No statement Or e)(presslon of opInion or any other matter herein contained IS, or IS to be deemedto be, dlreclly or Indirectly, an offer or the soliCitation of an offer to buy or sen any secunty referred to or mentioned The matter IS presented merely for the convenience of the subscnber While we beheve the sources of our information to be rehable, we In no way represent or guarantee the accuracy thereof nor of Ihe statements made herein Any action to be taken by the subscnber should be based on hiS own InvesligatlOn and information Delafield, Harvey, Tabelllnc, as a corporation and ItS officers or employees, may now have, or may later take, poslliOns or trades In respect to any secuntles mentioned In thiS or any future Issue, and such position may be different from any views now or hereafter 8)(pressed In thiS or any other Issue Delafield, Harvey, Tabellinc , which IS registered With Ihe SEC as an Investment adVisor, may give adVice 10 liS -Investment adVISOry and olher customers Independently of any statements made In thiS or In any other Issue Further information on any security mentioned herein IS available on request

'1l'lIi.\ alE D.. D.. ' S IiUiJ lIi.\ IRi 1E'1l' D..1E'1l''1l'1E1Rl 600 ALEXANDER ROAD, CN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC 1609) 987-2300 – -I— — – — – – – – – – – – – – – – – -October6,1989 I—' Having suffered through four weeks of relatively poor short-term action in September, the first week of October has indeed provided a welcome relief, i.e., an 80 point advance in four trading days to date. An interesting and perhaps simplistic way to explain away the poor market action in September is to re-examine a significant seasonal pattern in the stock market—the tendency toward a market decline in the month of September. Since 57 of all months since 1897 have been rising ones, the expectation would be a plurality of advances over declines. However, precisely the opposite is the case for September, which in 92 years, has produced 56 declines, including this year, and only 36 advances with an average drop of 1.28. The probability of such a pattern being due to chance, a chi-square test would reveal, is less than 1 in 1000. However, what has really changed Approximately two months ago on August 8, the DJIA posted a new high of 2699.17, and the advance/decline breadth line also posted a new high, confirming the advance. Since then, the DJIA has posted seven successive new highs, and, in each case, this has not been confirmed by a new high in market breadth. In spite of the 80 point advance in the Dow Jones Industrial Average this week, there continues to exist a negative short-term breadth divergence. Although recent improvement in market breadth is apparent, currently needing 1329 net advances over declines to confirm a continuance of the bull market, a confirmation is still needed. We are fast approaching the second anniversary of the October 19, 1987 major bear-market low. Since that time, the two-year-old bull market has to date scored an impressive advance of 59.5. Having posted a new record high yesterday, it would indicate that this .-l…. ..m…,ajor-cycle upswing has more room on the upside. That is not to say that the toppinog-I process. which recent readers of this letter have been alerted to, can begIn at any time. – Recent strength in the DJIA has, for the time being, canceled the potential top which previously existed. However, future market action CQuld broaden the potential top As we have noted from past history, a characteristic of major market tops is that they tend to form over a lengthy period of time. In any case, a significant top were it to develop, still would not be confirmed unless the DJIA were to break below the 2630 level. The clue to the direction of the stock market may still be found in terms of individual stock patterns. While most stocks have been in uptrends prior to this week's market action, minor deterioration had been confined to the formation of short-term tops in defensive-industry groups that have been market leaders over the past year, food stocks being an obvious group, and Coca-Coia, an excellent example. With the action of this week, however, many of these stock patterns have improved dramatically, as in the case of Coca-Cola, posting new record highs This improvement, coupled with the continued improvement in the cyclical, heavy-industry issues and in the energy sector, is, of course, encouraging. Short-term action, this week, continues to suggest that the current bull market has further to go, and the minor 3.38 decline in September was nothing more than a normal short-term downswing. However, it is also quite clear that upside activity, because of the longevity of the ongoing bull market is becoming more selective. This, together with rotational market leadership shifting into new areas (possibly from the consumer goods area into capital goods), should be monitored closely. Although this week's market action has been most impressive to date, we must not lose sight of the potential problems that were with us in the past few months, as the stock market goes higher. ROBER T J. SIMPKINS, JR. DELAFIELD, HARVEY, TAB ELL INC. Dow Jones Industrials (12 00) S & P 500 (1200) Cumulative Index (10/5/89) 278742 358.35 4962.20 AWTebh No statement or expression of opinion or any other matter herein contained IS, or IS to be deemed to be, directly or indirectly, an offer or the solicllatron of an offer to buy or sell any secUrity referred to or menboned The matter IS presented merely for the convemence of the subscnber While we believe the sources of our Information to be reliable, we In no way represent or guarantee the accuracy thereof nor of the statements made herein Any aellon to be taken by the subscnber should be based on hiS own In\'estlgatlon and Informaton Delafield, Harvey, Tabelllnc, as a / corporation and rts officers or employees, may now ha\'e, or may later take, poslliOns or trades In respect to any secuntles mentioned In thiS or any future Issue, and such poSition may be different from any \'Iews now or hereafter expressed In thiS or any other Issue Delafield, Harvey, Tabelllnc, which IS registered With the SEC as an Investment adVIsor, maY9IVe adVIce to Its Investment adViSOry and other customers independently of any statements made In this or In any other Issue Further Informallon on any security menboned herein Is available on request