Archive Press Releases

WRITTEN TESTIMONY to the

"Committee on Banking and Financial Services"

Of the U.S. House of Representatives,

September 14, 1999

by

Awi Federgruen

I am the Charles E. Exley Professor of Business and Senior Vice Dean of the Graduate School of Business at Columbia University. My areas of expertise are Management Science (including Statistics and Decision Models) and Operations Management.

In April of this year, I was approached by Counsel of Fernand Bodner et al. and Anne-Marie Benisti et al., plaintiffs against a consortium of French Banks in the Eastern District of New York of the United States District Court. At the request of Counsel for the Plaintiffs, I reviewed the 2^nd Report of the Matteoli Commission dated 31 December, 1998, which has been invoked in support of the position of the Republic of France as well as by the French banks. I prepared an affidavit in which my comments and analysis are confined to the second part of the Report, starting with the section entitled "Bilan provisoire estime de la spoliation concernant les avoirs deposes".

The 2^nd Report is mentioned as the principal current effort by the French government to "address the eternal debt France owes to the Jewish victims of the Holocaust". The Progress Report is described as a "massively thorough, detailed and fact-intensive summary of the results of the investigations". The existence of this report and the ongoing activities of the Matteoli commission are invoked as one of the principal reasons to oppose the above mentioned litigations . While it is unclear why these litigations would "implicate the legitimate interests of France," and why "such actions would be contrary to the efforts of the French government to achieve a comprehensive, thorough and complete resolution of all unresolved issues caused by the occurrence of the Holocaust in France", the unstated assumption is that the "massive thoroughness" of the Matteoli commission's investigations will ensure that the extent and magnitude of the "eternal debt France owes to the Jewish victims of the Holocaust" will be assessed accurately and in toto. It would appear that any information uncovered in the process of the above mentioned litigation could only assist the work of the Matteoli Commission and the French Government's sudden desire, fifty five years after the Holocaust, to "address the eternal debt France owes to the Jewish victims of the Holocaust".

The argument loses further credibility as the Defendants in the above mentioned litigations have in parallel stated that they cannot be held liable for robbing the Jewish community in France of their assets, at the most vulnerable point in their history, because, as Defendants claim, the banks were "only following orders" of the Vichy French authorities. This argument has, unfortunately, an all too familiar ring. Translated into German, it is the "Befehl ist Befehl", or "superior orders" argument which the Holocaust war criminals brought forward after the Second World War. Needless to say, this preposterous argument was rejected during the Nuremberg trials where it was consistently found that the circumstance that an individual or organization is following instructions of a supervising office or authority, provides no legal defense for violations of international law, whatsoever.

Moreover, what is at stake here in your Committee's hearings, or in the aforementioned litigations is not the prosecution of violations of international criminal law, whether committed entirely voluntarily or not, but an organized attempt to restore as many of the illegally appropriated assets to their rightful owners or their heirs, as possible. While the heinous crimes of the Nazi regime and their French collaborators can, unfortunately, not be reversed, it is of course eminently feasible to restore assets to their rightful owners, once identified as such.

Juxtaposed with the above claims, the argument that the work of the Matteoli Commission would prelude the necessity or even the mere utility of litigations in this country, loses any form of credibility in and of itself. However, since I am neither a lawyer, nor a historian and since my views and conclusions regarding the above historical and legal issues add little or no value to what has been or will be said by others, far more qualified in these areas, I will confine my remarks to an assessment of the contents of the Matteoli report.

This assessment is preliminary in that it addresses only a subset of the problems associated with the report, and, as mentioned, my comments are confined to the second part of the report, starting with the section entitled "Bilan provisoire estimé de la spoliation concernant les avoirs deposes". It is this part which attempts to arrive at an estimate of the magnitude of the assets of Jewish citizens of France, which were withdrawn, alienated, or blocked during the Second World War.

Principal observations

The "Report" distinguishes between three categories of assets belonging to Jewish citizens of France and despoiled by the German occupiers and Vichy regime. (II) Deposited assets, withdrawn or alienated. (III) Assets blocked which remained on deposit in the banks. (IV) Safe deposit boxes. The "report" claims that in terms of the assets in category (II) approximately the same amount was "returned" (presumably to the rightful owners or their heirs,) as was "withdrawn" or "alienated". Leaving the accuracy of the assessments regarding (II) aside, momentarily, the committee is only in the preliminary stages of assessing the magnitude of the assets in categories (III) and (IV). While the assets in category (III) were "in principal" "released", after the liberation, by a series of government acts, the Matteoli report itself admits that restoration of these assets to their rightful owners occurred only if the following two conditions were satisfied and subject to the following caveats:

(1a) "By this measure, the owners were therefore given free access to the balance of their accounts, and to their securities, if they had not been sold and if they were not under provisional administration. On the other hand, with the aim of protecting the value of currency, gold and foreign currency and securities remained blocked for all clients" (p. 218, English translation).

(1b) "The person whose assets had been despoiled, should have had the balance available of his deposited assets returned to him, on condition that he was present or represented."

2. As to (1a), nowhere does the report address the extent to which assets were "sold", or "under provisional administration". Nor does the report address the volume of holdings in gold and foreign currency. It stands to reason that the latter represents a large percentage of the holdings of the Jewish population, knowing all too well, that in the face of war and persecution, it needed to maintain its savings in the most liquid and portable form possible, i.e. in gold and foreign currency. The qualification under (1b) is equally if not more restrictive. The deported part of the Jewish population was, by the end of the war, either murdered by the German occupier and its French accomplices or found itself in D.P. camps in other countries and it was therefore not in a position to be present or represented, so as to take possession of their despoiled assets.

3. Most importantly, perhaps, the Matteoli report admits on p. 218, that for each restitution of securities placed under the Estate department's provisional administration, a "delivery report drawn up in the presence of both parties" was required, but that "not a single copy of such reports has been found in the archives". Similarly, the Matteoli report admits that while, in principle, all blocked assets were "released", upon the country's liberation "their release does not appear to have left any traces". The Matteoli report continues to state that its future activities in 1999 will in part be devoted to assessing the extent to which restitution of blocked assets in fact took place, an effort which the Committee characterizes as "most difficult, since their release does not appear to have left any traces."

4. Noting, in addition, that by the Matteoli report's own estimates, the magnitude of the assets under (III) (blocked assets which remained on deposit) is close to three times that of (II), the implication of the report is therefore that the restitution of the majority of the assets is in question (approximately 5 billion Francs). The Matteoli report, itself, is therefore in sharp and unambiguous disagreement with the claim by the French banks, according to which "these analyses reflect a payment during that period of approximately 95% of the funds and other assets that had been taken during the war".

5. The second part of the Matteoli report is written in a highly disorganized manner. Rather than specifying and justifying a clear methodology to estimate the magnitude of various asset pools or the number of potential claimants to these pools, it lists various partial calculations, often of quantities not used in the remainder of the report. The assumptions underlying various calculations are often flawed, as some of the examples below will indicate. There is no clear correspondence between numbers stated in summary tables and those resulting from corresponding analyses or discussions. For example, p. 217 states that the magnitude of the assets in category (II), i.e. assets subject to withdrawals or alienation, is in the amount of 2,333 billion francs, and that in the period 1944-54, 2.4 billion francs was returned. On the other hand, on p. 221, in section IV-4, the report claims that for assets withdrawn by order of the Vichy government alone, by 1951, i.e. 3 years before 1954, 3 billion francs were returned in a total of 4,300 cases.

6. Many of the numbers stated in the report are entirely unsubstantiated, either by an estimation procedure or by a direct reference to other reports or archival courts. For example, the summary statement on p. 217, pertaining to an estimated statement of spoliation of deposited assets, states that the number of deported Jews with blocked deposited assets is no more than 7,400, less than 10% of all deported Jews. This number fails to be substantiated and appears incredulous. The same applies to the stated figure of 53,000 Jews and non-Jews with deposited blocked assets.

7. The Committee appears to be satisfied with exceptionally wide estimate ranges, in which the upper limit of the estimate is no less than 10 times the lower limit estimate. For example, the estimate for the aggregate volume of blocked bank deposits not subject to withdrawal or alienation ranges from 50 to 500 million francs. (See the original French version on p. 95; the English translation conspicuously quotes the lower limit of 50 million francs as an exclusive estimate.) It appears that the Committee is content with their current estimate ranges in which the upper limit is no less than ten times the lower limit, as their future work plans do not include any attempts to replace the current ranges by a meaningful and implementable point estimate and associated confidence interval. No serious professional statistical estimation methodology would conclude with such meaningless estimate ranges.

8. The Committee does not begin to address how the nominal value of the unsubstantiated, blocked and alienated funds, in French Francs of 1941 is to be converted into equivalent sums, close to 60 years later. To maintain purchasing power parity alone, one $ owned in 1941 would need to be compensated by perhaps as much as $10, if the Consumer Price Index is to be used as the deflator. (For example, in the US, the Consumer Price Index in 1996 was 8.62 times its value in 1946.) In addition, it needs to be recognized that any savings in 1941 would have been invested, eg. in fixed income or equity markets, resulting in yields considerably larger than those required to offset inflation, i.e., to maintain purchasing power parity, alone. There is no unique, unambiguous way to find reasonable conversion factors over a 60 year time interval. (An additional complication arises from the fact that many of the Holocaust survivors or heirs to Holocaust victims have moved to different countries, requiring an appropriate currency conversion resolution). Serious studies are, however, underway to establish such conversion factors for other assets held by Holocaust victims, eg. by the holders of insurance policies. The Matteoli report is conspicuously silent about this entire issue, leaving the reader with the impression that the Committee views outstanding balances in 1941 French francs as the appropriate measure to arrive at current obligations and restitutions.

9. Much of the second part of the Matteoli report is devoted to arrive at an estimate of blocked bank deposits, the main component of the first item under category (III) on p. 217. (Recall this component is estimated to amount to anywhere between 50 million and 500 million francs.). Two approaches are pursued, but both are fundamentally flawed:

a)  An estimation procedure based on a sampling technique;

b)  an estimation procedure based on disaggregation from known national aggregate figures. We critique both procedures separately under 10. and 11.

10. The sampling technique employed suffers from a number of methodological flaws and its results should therefore be dismissed. It is my understanding that shortly after the German occupation all French banks were required to send out a form to all account holders requesting self identification as Jewish or "Aryan". Subsequently, on December 20, 1941 the French banks were required to report which of their blocked accounts belonged to natural persons and legal entities "considered to be Jewish," presumably on the basis of the information collected via the aforementioned forms. It appears that on that date a total of 67,962 blocked accounts were reported as "considered to be Jewish". To assess the magnitude of the balances in these accounts (on December 20, 1941) all accounts in the above population of 67,962, whose names start with the letters BA were "pulled", and a complete distribution of account balances was tallied. Here are some of the major problems associated with the sampling technique:

• The number of blocked accounts belonging to Jews is likely to be considerably larger than the 67,962 who decided to self identify themselves as Jewish to the Nazi occupants and their French accomplices, engaged in their implementation of the "Final Solution". It stands to reason that a large percentage of the remaining blocked assets belonged to Jews as well, given the large Jewish representation among opposition parties and underground movements or other segments of the population that were particularly vulnerable to the blocking measures.
  
• It is unclear why the full population of 67,692 account balances could not have been tallied with the help of standard statistical software packages. Even if a full tally was deemed to be too expensive, or too time consuming it is unclear how the Committee decided to use a sample of 1,000 accounts only.
  
• Most importantly, reasonable sampling techniques would draw randomly from the total population instead of on the basis of a given pair of starting letters of the corresponding names (BA). For one, the procedure begs the question why this specific pair of letters was chosen?! In addition, even if the pair "BA" was randomly chosen from all 26x26 = 676 possible such pairs, there are likely to be significant biases arising from this sampling choice. For example, Jewish names are often associated with specific professions and socio-economic backgrounds, even if the distribution of geographic origins (Central Europe, Southern Europe and North Africa) in the sample - clearly an attribute of no direct relevance to the estimation problems at hand is reasonably close to that of the entire Jewish population, as the Matteoli report asserts on p. 246.

11. The disaggregation technique as described in section 11-2 employs the following logic:

• Since the Jewish population in France in 1941 represented approximately 0.7% of the total population, it is assumed that the total value of bank accounts held by Jews can be estimated as 0.7% of the national aggregate. For individual deposits, for example, this proportionality assumption results in an estimate of 1,278 million francs. The assumption is entirely unreasonable, resulting in severe underestimations, and this for at least the following two reasons:

11a) Jews in 1940 France used the banking system as a vehicle to hold their liquid funds, in much greater percentages than the overall, largely agrarian and rural, French population.

11b) The above proportionality assumption is based on the "implicit" belief that the average account size within the Jewish population equals that of the entire French nation. This assumption is equivalent to estimating the account balances among Manhattan and Beverly Hills residents, for example, by multiplying aggregate US figures with the percentage of the US population residing in these two unrepresentative locations.

12. The nationwide Jewish balance estimates are multiplied with the percentage of all nationwide Jewish and non-Jewish accounts belonging to owners in the "occupied zone". This disaggregation step is again subject to at least three fundamental flaws, each of which has the impact of severely underestimating Jewish holdings in the occupied zone.

12a) The disaggregation step is presumably based on the assumption that the only Jews subject to the blocking measures were those residing in the "occupied North of France". Similar blocking measures prevailed, however, in the non-occupied part of France as well where, ironically, they were initiated as a formal statutory matter at an even earlier date. See Sarraute and Tager "Les Juifs sous l'occupation", Centre de Documentation Juive Contemporaire (1982), p. 41: The Law of 26 April, 1941 permitting the blockage of certain bank accounts.) Thus, the full nationwide Jewish balance estimates should be used rather than any regional fraction thereof.

12b) Even if point 12a) is to be ignored, the percentage of Jews living in the occupied zone was likely to be significantly larger than the corresponding percentage in the total French population, as the Jewish population resided primarily in the larger cities in the occupied zone.

12c) Even if points 12a) and 12b) are to be ignored, bank accounts in the northern or "occupied" part of France with its many commercial and industrial metropoli including Paris itself, have a significantly different profile than the one pertaining to the entire country. This is likely to result in another significant downward bias..

• Even if the above estimates of 1,278 million francs for Jewish owned individual deposits were correct, and even if the above disaggregation step could be justified, the aggregate of Jewish owned, blocked individual deposits should be estimated as 67% x 1,278 million = 856 million francs, since presumably 67% of all French individual deposit accounts resided in the "occupied zone". Instead this quantity is estimated as 313 million francs. Similar gross "calculation errors" appear to apply to other aggregate account balances as well.

13. In conclusion, the Republic of France and the banks invoke the Matteoli commission report, dated 31 December, 1998 as its principal current effort to address the internal debt France owes to the Jewish victims of the Holocaust. The Republic of France characterizes this report as "massively thorough". In fact, the second part of the report is written in a highly disorganized manner. Rather than specifying and justifying a clear methodology to estimate the magnitude of various asset pools or the number of potential claimants to these pools, it lists various partial calculations, often of quantities not used in the remainder of the report. The assumptions underlying various calculations are often flawed and fail to conform with state-of-the-art standards. Important questions are entirely ignored, eg. the conversion of nominal "debts" to Holocaust victims in French Francs of 1941, to equivalent US $ in the year 1999, see 12., above. Finally, the Republic of France and the French banks misrepresent the findings of the Matteoli commission, whatever the latter's validity may be.