by Edwin Black
Watson compromised—in a way. With his consent, Dehomag adopted a shareholder resolution for “an eventual dividend to be declared for the years 1935, 1936 and 1937.” When it was, Heidinger would be paid his long awaited bonus, less all his advances, of course. In the meantime, Watson’s many outside advisors would provide the written opinions about how much profit was legally permissible to declare under existing German law without incurring confiscatory taxes and mandatory loans to the Reich. To assuage a nervous Heidinger, Watson agreed to provide yet more advances, RM 7,000 monthly for the remainder of 1938.107
But Heidinger was impatient. While he had appealed the tax decision, he did not expect to prevail. Soon, Heidinger would have to pay a huge assessment. Dehomag’s books reflected one multi-million mark record after another—1938 alone would yield RM 2.39 million in conceded profits even after IBM applied various intra-company devices.108 Yet Watson still delayed any decision on declaring a profit.
Finally, in late November 1938, just days after Kristallnacht, a furious exchange of correspondence between New York and Berlin escalated into a stubborn standoff over dividing the money.
The squabbling culminated with Heidinger implying that Watson was involved in defrauding the Reich tax authorities. In a long, rambling and sarcastic five-page letter to one of Watson’s Berlin attorneys, Heidinger openly conceded his stock was a sham. Referring to his so-called “preferred shares in Dehomag,” Heidinger declared, “My company shares are no real preferred shares, if for instance the Tabulating Division would yield no net profit, while the remaining divisions would earn a net profit of say five percent, on my shares, I would not obtain anything and the remaining five percent are therefore not preferred in that case but disadvantaged.”109
Heidinger’s letter repeatedly insisted the bogus share arrangement might be viewed by the authorities as a scheme “flatly to evade paragraph 3 of the law.” He invoked strong words, uncharacteristic of IBM’s usual ambiguity. At one point, he referred to “a tax liability evaded by abnormal measures.” The word “evade” was used repeatedly, as in “tax evasion.” Heidinger even added an unsubtle hint of criminality, writing, “But by no means must we expose the Dehomag to the risk of a penal prosecution.” As was his style, he flamboyantly concluded his pejorative missive “with renewed hearty thanks.”110
Watson sought help from Price Waterhouse. But the prestigious accountancy firm could only conclude Dehomag’s finances were in supremely profitable condition and that Heidinger deserved his bonus. In its lengthy thirteen-page single-spaced analysis, dated December 30, 1938, Price Waterhouse declared: the only question is when and how much to pay Heidinger. Moreover, warned Price Waterhouse, if Heidinger insisted on selling his shares, the value of that stock—real or not—was far greater than when the original merger took place in 1934. Using rigid principles of valuation, Price Waterhouse examined the pluses and minuses of the German political and tax environment, and the problem of blocked bank accounts. The firm concluded that each share of Dehomag was actually “worth more to a purchaser in Germany, than to a resident abroad.” The report underlined the words “in Germany.”111 For Watson, this meant that his shares were now actually less valuable than Heidinger’s.
Indeed, Price Waterhouse asserted, Dehomag by any measure had only become more valuable. The net worth of the company had essentially doubled from its RM 7.7 million total investment in 1934 to more than RM 14 million. Annual earnings were about RM 2.3 million, a 16 percent return on net assets.112
At the same time, more bad news came. Dehomag was supplying machinery and spare parts to IBM for resale throughout Europe. IBM in turn merely credited Dehomag’s loan balance account. Frustrated and defiant, Dehomag managers in mid-December 1938 unilaterally began terming those shipments “exports.” This triggered the Reich’s rule requiring actual foreign currency payment, which Dehomag obtained by debiting IBM’s precious few dollar accounts in Germany.113
On January 3, 1939, IBM’s Geneva Controller J. C. Milner mailed Watson a long, detailed letter explicating the adverse Price Waterhouse report, searching for silver linings, parsing Heidinger’s contract language, and ultimately trying to construct loopholes around the inevitability of either paying Heidinger dividends or buying part of his stock. Milner conceded that buying just one of Heidinger’s shares would expose the subsidiary as American-controlled.114
Milner explored all the possibilities. “If he [Heidinger] died and the stock was offered to IBM, in accordance with his contract, the higher book value combined with the earnings of the company would probably force a high valuation of the stock,” asserted Milner. Maybe the company could pay the elderly Heidinger in ten annual installments? Could Dehomag purchase Heidinger’s stock with blocked marks as an internal obligation? Milner offered a range of options, none of them promising.115
It seemed to be a no-win dilemma for IBM. Purchase of Heidinger’s stock was out of the question, asserted Milner, because no one could predict what the Reich economic and taxing authorities would do. On the other hand, once dividends on the 1935-1937 period were formally paid to Heidinger, he would next ask for dividends for 1938. It would continue annually even as the company’s value escalated.116
Clearly, money was a pressure tactic IBM could use. Heidinger was receiving a monthly allowance of RM 7,000 for all of 1938. Milner had some weeks earlier reminded IBM attorneys in New York, “the last payment on this account will fall due in December. It will then be necessary for a decision to be made regarding the year 1939.”117
Heidinger was being squeezed. Not only was he liable for a RM 90,000 tax, but because of the protracted reporting delay, German tax authorities had added a mandatory loan to the government, made retroactive for the three years 1934-1937, and that loan totaled RM 151,000. He could never afford that without help from the company. Watson understood that, and cut off Heidinger’s RM 7,000 monthly advance.118
In a March 13 letter, Rottke implored Milner to advance Heidinger the money needed for the mandatory loan.119 Heidinger was clearly desperate. Tax monies would be due within a matter of weeks. He had accommodated Watson all these years. Now he needed help.
Watson was unmoved. On March 15, the day the Nazis smashed into the remainder of Czechoslovakia, Milner calmly answered Rottke, expressing regret for a “very awkward condition.” But in fact, insisted Milner, it was Heidinger who had insisted that dividends be paid. If now the taxing authority had imposed mandatory loans, that was Heidinger’s problem. Indeed, IBM attorneys in Berlin had carefully studied current regulations and determined that IBM had actually advanced monies above the legal limit. “Therefore,” asserted Milner, “it is Mr. Heidinger who has received too much money, and it is he who should make arrangements to invest the surplus with the Loan Stock Bank.”120
In describing the mess to the IBM NY officers, Milner caustically noted, “We cannot be blamed if Mr. Heidinger’s own government will not let him draw adequate cash dividends. On the other hand, this increases the hazard of his offering to sell us some of his stock.”121
In the meantime, IBM was negotiating with the subsidiary’s two other managers, Hummel and Rottke, over the profit sharing plan for Dehomag’s activities outside of Germany. On March 21, six days after Czechoslovakia had been seized, even as Poland, Lithuania, and other countries were being actively threatened with German invasion, IBM European troubleshooter Harrison Chauncey dashed a short note off to Phillips about the bonus terms for “new territories” to be handed to Dehomag. “I wonder,” Chauncey asked, “if the further changes in the German political situation require any consideration of this subject at this time?” Phillips in New York scrawled a note back, “Considering present changes in the map of Europe don’t you consider it best to wait?”122
It was no longer just Austria and Czechoslovakia. Clearly, other nations would soon come under Dehomag’s sphere of influence. IBM was trying to plan ahead.
* * *
BRINKSMANSHIP WAS Watson’s specialty. First he
instructed Holt not to go to Berlin to participate in a scheduled shareholders meeting. Hence, no decision could be voted on Heidinger’s request. As each day passed, Heidinger’s financial situation worsened.123
Then, on March 31, 1939, Watson cabled Holt: “Loan Heidinger 150,000 marks to pay Loan Stock Account and also authorize you to vote for payment of 8% dividend, you to invest our dividend money in real estate.”
Under German law, 8 percent was the legal limit IBM could pay without incurring additional taxation. The 8 percent dividend was to be paid monthly just as the advance was. But 8 percent would total RM 3,500, just half of what Heidinger needed to pay his bills and half the 16 percent return identified by Price Waterhouse.124 Heidinger needed RM 7,000 per month. He was fed up with IBM and Watson.
APRIL 26, 1939
Thomas J. Watson
President of the IBM
New York
Dear Mr. Watson!
As you know, up to the end of last year, I received a monthly payment of RM 7000—as an advance on account of dividends…. these payments have been stopped since Jan. 39 … since that date, no shareholders meeting took place and therefore a corresponding resolution could not be formed.
A meeting has been called for April 11 … Mr. Holt replied … “it is not convenient for him to come Berlin” and that he acts solely in the capacity of the chief stockholder, the IBM….
April 14, Mr. Rottke wrote to Mr. Milner … saying among other words: “I seriously fear that Mr. Heidinger gets in economic difficulties … therefore I beg you kindly to discuss this item with Mr. Watson in Paris … Today, Mr. Rottke informed me that he received a letter of Mr. Milner … “to advance to Mr. Heidinger on account of dividends for 1939, a sum equal to eight percent of his capital share in Dehomag. (That means RM 3,500). This may be advanced monthly … and can be ratified at an eventual meeting of the partners.” …That means that the IBM either does not like my partnership or at least that it does not attach great value to maintain my partnership in the Dehomag.
Unnecessary to say how sorry and how deeply depressed I feel about such an attitude, which in all probability ends my partnership … I herewith offer my shares … in the Dehomag to the IBM … negotiate with me about the purchase price … accept the transfer of the shares to IBM.
I would be very happy and highly appreciative if the personal relations which have been created during the past 29 years between me and the different gentlemen of the IBM … and between you and me will not be changed… Again expressing my deepest regret, I beg to accept my personal regards and remain
very sincerely yours,
Willy Heidinger125
Rottke openly conceded the contract between IBM and Heidinger had “been made under an unlucky star, [and] appears to be the source of all evil.” But he nonetheless warned Watson again that if Heidinger’s shares were transferred to a foreign source Dehomag would probably not be permitted “the use of the word Deutsche (German) as an enterprise recognized in Germany as German.”126 That disaster had to be avoided at all costs. To IBM’s doctrinaire German managers, including Heidinger, Dehomag represented far more than just a profit-making enterprise. To them, Dehomag had the technologic ability to keep Germany’s war machine automated, facilitate her highly efficient seizure of neighboring countries, and achieve the Reich’s swiftly moving racial agenda. If IBM’s subsidiary were deemed non-Aryan, the company would be barred from all the sensitive projects awaiting it. Hitler’s Germany—in spite of itself—would be deprived of the Holleriths it so desperately required.
From Watson’s point of view, Germany was on the brink of unleashing its total conquest of Europe. IBM subsidiaries could be coordinated by Dehomag into one efficient continental enterprise, moving parts, cards, and machines as the Reich needed them. The new order that Hitler promised was made to order for IBM.
In July 1939, Watson arrived in Berlin to personally mediate with Heidinger. A compromise would be necessary. The stakes were too high for the Nazis. The stakes were too high for capitalism. But it was the Germans who gave in, deferring on Heidinger’s demands for a few months under terms Watson dictated. Watson now controlled something the Third Reich needed to launch the next decisive step in the solution of the Jewish question, not just in Germany—but all of Europe. Until now, the fastest punchers, tabulators, and sorters could organize only by numbers. The results could then be sorted by sequentially numbered profession, geographic locale, or population category. But now Watson had something new and powerful.127
He had the alphabetizers.
VII. DEADLY COUNT
ON MAY 17, 1939, GERMANY WAS SWEPT BY 750,000 CENSUS takers, mainly volunteers. They missed virtually no one in the Greater Reich’s 22 million households, 3.5 million farmhouses, 5.5 million shops and factories. Teams of five to eight census takers fanned out through the big cities such as Berlin, Frankfurt, Hamburg, and Vienna. Towns and villages were divided into districts of thirty homes with one census taker assigned to each. Some 80 million citizens in the Greater Reich, including Germany, Austria, the Sudetenland, and the Saar, would be classed according to their ancestry.1
There was little question to the world that the May 1939 national census was racial in nature. New York Times coverage of the mammoth project made clear that this census would “provide detailed information on the ancestry, religious faith and material possessions of all residents. Special blanks will be provided on which each person must state whether he is of pure ‘Aryan’ blood. The status of each of his grandparents must be given and substantiated by evidence in case of inquiry.”2
Certainly, by May 1939, virtually every “practicing Jew” had been registered, surveyed, numbered, and sorted numerous times in a series of overlapping, often disjointed, campaigns. The purpose of the 1939 census was to identify the so-called “racial Jews” in Germany proper, add Jews of any definition in the new territories of the expanded Reich, and locate each individual before being ghettoized or subjected to other action. Indeed the ghettoization decrees had begun that very month. In addition, Germany was preparing for all-out war and without the census, it could not identify exactly where all its draftable men were, and which women would step into their economic shoes once mobilized.3 As such, the census was vital to Hitler’s two-front war—one against the Jews, and one against all of Europe.
Understandably, Dehomag’s 1939 undertaking dwarfed anything it had attempted before, including the 1933 Prussian census. Months of intensive training, conducted in thousands of sessions, prepared legions of volunteers for the critical mission. Police and their auxiliaries were mandated to support the count “with all their powers” and “to function as census-takers in difficult and confusing residential areas,” according to official regulations.4
The additional Hollerith machinery assembled was massive: 400 electrical key punches, 10 gang punches, 20 summary punches, 300 key punch verifiers, 70 sorters, 50 tabulators, 25 duplicators, and 50 D-11 VZ tabulators. The Reich had imposed seemingly impossible target delivery dates for November 1939. So to increase speed, Dehomag’s engineers converted their versatile D-11 calculating tabulator into a pure counting machine dubbed the D-11 VZ. The improvised device could process 12,000 60-column punched cards per hour in sixteen counters and then precision-punch its own summaries onto 80-column cards. Eighty million cards were actually used.5
A special envelope containing a so-called Supplemental Card was created. This all-important card recorded the individual’s bloodline data and functioned as the racial linchpin of the operation. Each head of household was to fill out his name and address and then document his family’s ancestral lines. Jews understandably feared the newest identification. Census takers were cautioned to overcome any distrust by assuring families that the information would not be released to the financial authorities.6
But it was not German taxing agencies that were the most eager for the new data. It was the Nazi Party structure and Reich security forces seeking to locate additional Jews and other u
ndesirables. Indeed, the final data was intended to help comprise a single national register for the entire Greater Reich. Each card carried a single column coded for descent, designed into the card prototype long before the census was launched. A letter from the Order Police to the Ministry of the Interior at the end of 1938 explained: “This column on the registration card is included to be filled out at the right time. That time should come in May of next year during the population, occupation, and company census. The regular questionnaire will be supplemented by an additional card. This card will include the question of whether the person had any fully Jewish grandparent. Survey results will then be evaluated using this registration card.”7
The 25 million Supplemental Cards—one for each household—represented a virtual doubling of census files. To cope with the volume and still meet deadlines, the census tabulation was divided into two operations. First, each special envelope containing a Supplemental Card was labeled to correspond to the household’s general questionnaire, along with the district and municipality of origin. Then local officials, generally the police, affixed the letter “J” to both the questionnaires and cards of all Jewish families.8
The words “Do Not Send Directly to the [Berlin] Statistical Bureau” were printed on every envelope. Instead, both the general questionnaire and its companion special envelope were sent to the regional statistical offices for the tedious quality control procedures. Did the envelopes match up to the questionnaires? Were Supplemental Cards containing racial data and the general questionnaires filled out completely? Just preparing the 25 million census forms and 25 million Supplemental Cards for processing required a behemoth manual operation. Once approved, the questionnaires and cards were transported to Berlin and separated. The Supplemental Cards were sorted into three groups: non-Aryans, “higher educated people,” and all others. These were then tabulated to yield the racial data.9
