Corrado Gini

Motta di Livenza (Treviso), 23 May 1884 — Rome, 13 March 1965

This website is not the scientific journal. Metron – International Journal of Statistics is published by Springer: link.springer.com/journal/40300. This page is the successor to the biography that stood on the journal’s former website, rewritten with its sources.

Statistician and demographer. In 1920 he founded Metron – International Journal of Statistics, now published by Springer (its history). He was the first president of Italy’s Central Institute of Statistics. His name is attached to the measure by which inequality is most often quantified — and to a relationship with the Fascist regime that historians have examined in detail.

The Gini index, with the right date

Look up when the Gini index was introduced and almost every source will say 1912. That is a simplification worth unpicking, because two distinct works are involved.

• In Variabilità e mutabilità (Bologna, 1912) Gini introduces the mean difference and the conceptual framework of variability and mutability. This is where the idea originates.
• In Sulla misura della concentrazione e della variabilità dei caratteri (1914) he introduces the concentration ratio — the formulation now generally called the Gini coefficient or Gini index.

MacTutor attributes the Gini coefficient explicitly to the 1914 work. The Journal of Economic Inequality, on the other hand, republished extracts from Variabilità e mutabilità under the title The origins of the Gini index, placing the origin there.

This is not pedantry. It is precisely the kind of imprecision that propagates for decades because nobody checks it. The two dates describe two different things: the idea, and the formula.

What it measures. The Gini index compresses into a single number between 0 and 1 how concentrated a quantity — typically income — is in few hands. Zero is perfect equality; the maximum is concentration in a single holder — on a finite population of n units that value is 1 − 1/n, with 1 as the limit.

What it does not measure, and why that matters. Two very different distributions can share the same Gini index. A rising index does not tell you where inequality grew — at the top, at the bottom, or in the middle — and it says nothing about absolute income levels. It is a summary, and like every summary it discards information. It is a good place to start a question, not a place to end one.

Career

Law degree at Bologna in 1905; qualification to teach statistics in 1908. He taught at Cagliari from 1909 — full professor from 1910 — at Padua from 1913, and at Rome from 1925.

In 1920, in Padua, he founded Metron, as both editor and owner. In 1926 he became the first president of the Central Institute of Statistics — today’s Istat — a post he held until 1932, when he resigned (the resignation was submitted in late 1931).

He went on to found the Italian Committee for the Study of Population Problems (1929), the journal Genus (1934), and in 1936 the Faculty of statistical and actuarial sciences in Rome, of which he became dean — described by the Dizionario Biografico degli Italiani as «the only one in Italy and in Europe» at the time.

(Two sources disagree here, and it is worth knowing: the DBI dates the Faculty to 1936, while Sapienza’s Department of Statistical Sciences writes that the Faculty — «the first of its kind in Italy» — was founded by Gini «in the 1920s». The two readings reconcile if one notes that the DBI places the creation of the School of Statistics in 1927, from which the Faculty descends: the 1920s are the origin, 1936 is the Faculty.) He was president of the Italian Statistical Society from 1941 to 1944, and again from 1949 until his death. (Note: Gini did not found the society — it was established on 15 January 1939, and its first president was Gaetano Pietra, 1939-1941.)

Gini and Fascism

This section cannot be skipped, and it cannot be handled by insinuation either.

The established facts. Gini was appointed to the presidency of the Central Institute of Statistics by Mussolini, and held it from 1926 to 1932. He was president of the Italian Society of Genetics and Eugenics (1934) and represented that society at the Third International Congress of Eugenics in New York in 1932. His role in Italian and international eugenics is documented.

He did not sign the Manifesto of Racial Scientists of 1938. The ten signatories are documented and his name is not among them. It was signed by Franco Savorgnan, who had succeeded him as president of the Central Institute of Statistics — a fact that touches the institution but not Gini personally. It is worth stating plainly, because the mistaken attribution circulates.

The historiography. The standard reference is Francesco Cassata, Il fascismo razionale. Corrado Gini fra scienza e politica (Carocci, 2006), the first study to work systematically through Gini’s private archive. Cassata’s account is harder than the one found in popular biographies: Gini was regarded as “the Duce’s scientist”, at least until his marginalisation in 1932; his demographic and sociological theories contributed to the formation of Fascist culture; his positions on eugenics were compatible with those of the regime; and his cyclical theory of nations supplied a “scientific” justification for Fascist colonialism and militarism.

Paradoxically, it was his belief in the autonomy of science from politics that led to his marginalisation. For Gini, politics should conform to science, not the reverse. That conviction led him to defend the statistical institute against interference from the regime — and, at the same time, to a militant conception of science in the service of Fascist demographic and imperial policy.

Some sources report that from volume XIV (1940) the editorial board was removed from Metron‘s title page in order to avoid expelling its Jewish members as the racial laws required. This claim appears in the journal’s own centenary editorial: it is an internal source — the journal describing itself — and should be read as such.

After the war. In the summer of 1944 Gini was subjected to purge proceedings. On 6 November he had to give up teaching, the deanship of the Faculty of Statistical Sciences, and the presidency of the Italian Statistical Society. He was acquitted of the most serious charges on 24 January 1945, but suspended from all academic posts and deprived of his salary for a year. The proceedings closed on 17 December 1945 with a finding of no case to answer. He resumed his duties in 1946.

A note on how this page is written

The old Metron website carried a biography of Gini written by the journal about its own founder, and it is that page which Wikipedia, MacTutor and several universities still link to.

Reproducing it would have been easier. But a site concerned with measuring things properly cannot make an exception for the man it owes its name to. What can be done — and what has been attempted here — is to separate established facts from interpretations, to say who argues what, and to leave the judgement to the reader.

Sources

Domande frequenti

When was the Gini index introduced?
It depends what is meant. The underlying concepts — mean difference, variability, mutability — appear in Variabilità e mutabilità (1912). The concentration ratio in its modern form appears in Sulla misura della concentrazione e della variabilità dei caratteri (1914). MacTutor attributes the coefficient to the 1914 work.

What does the Gini index actually measure?
How concentrated a quantity is in few hands. It runs from 0 (perfect equality) to 1 (a single holder owns everything).

What does the Gini index not tell you?
Where the inequality arose: two very different distributions can share the same index. And nothing about absolute income levels — a poor country and a rich one can have the same Gini.

Did Corrado Gini sign the 1938 Manifesto of Racial Scientists?
No. The ten signatories are documented and Gini is not among them. It was signed by Franco Savorgnan, who had succeeded him as president of the Central Institute of Statistics.

What is this website’s relationship to the journal Metron?
There is no editorial relationship. Metron – International Journal of Statistics is published by Springer at link.springer.com/journal/40300. This site hosts the journal’s historical domain and documents its history.