Analysis of submissions by types of measures.

(a) Constitutional Amendments: Reference has already been made to the fact that a compulsory referendum existed in all states. except Delaware before the adoption of the initiative and referendum by any state, and to the fact that this compulsory referendum has continued for such measures. The compulsory referendum upon constitutional amendments through legislative submission of such amendments presents to the voters of some states without the initiative and referendum a large number of measures to be voted upon at each election. Not only this, but in a state like California, the legislative submission of constitutional amendmens was responsible for the submission of numerous measures to the people before the adoption of the initiative and referendum. In Louisiana which does not have the initiative and referendum, there were submitted by the legislature, in the ten biennial elections between 1900 and 1919, one hundred and thirty-two constitutional amendments of which one hundred and six were adopted. In 1912 the legislature proposed to the voters of Louisiana nineteen constitutional amendments and in 1914 and 1916 eighteen constitutional amendments for each year.

California adopted the initiative and referendum in 1911, the first measures under which were submitted in that state in 1912. In 1911, however, the legislature of California submitted 23 proposed constitutional amendments to the voters of that state and in each of the years 1906 and 1908 the legislature of California submitted fourteen measures to the voters. The fact of this frequent submission of constitutional amendments before the adoption of the initiative and referendum in California must be taken into consideration in considering the very large total of submissions since the introduction of the initiative and the referendum in that state.

Referring to the table of total submissions on page 103, it will be noted that in the states having the initiative and referendum seven hundred and seventeen measures have been submitted to a popular vote. Of these measures four hundred and twenty-nine were proposed constitutional amendments upon which the referendum is compulsory. That is, in order to change the constitution the measures must be submitted to a popular vote. Of the four hundred and twentynine constitutional amendmens submitted to a popular vote, two hundred and sixty-nine were proposed by the legislature and one hundred and sixty by the initiative. A large number of these proposed constitutional amendments relate to matters of relatively small importance and the popular submission in the larger number of these cases was made necessary by the fact that if the changes were to be made at all they must be made by constitutional amendment. That is, more than four out of each seven of the proposals submitted to the people in states having the initiative and referendum were submitted in large part because of the fact that detailed state constitutions made it necessary to employ a compulsory rather than an optional referendum. A very large number of the constitutional amendments proposed by the legislature were ones upon which a

popular vote would in all likelihood not have been demanded had the necessity for such a vote depended upon popular petition.

In 1914. forty-seven measures were submitted to a popular vote of the people of California, but of these forty-seven, thirty were proposed constitutional amendmens and twenty-two of these proposed constitutional amendments were submitted by the legislature. Referring now to the elections in which the largest number of measures has been submitted to a popular vote (Table on page 103), the optional referendum, that is a referendum not necessary for the adoption of the measure, was applied to but seventeen of the forty

seven measures.

In the Oregon election of 1910, thirty-two measures were submitted to a popular vote, of which eleven were constitutional amendments. In the Oregon election of 1912, thirty-seven measures were submitted to the popular vote, of which fourteen were constitutional amendments. In the Oregon election of 1914, twenty-nine measures were submitted to the people, of which nineteen were constitutional amendments. The statements here made are perhaps sufficient to indicate that to a very large extent the measures submitted in initiative and referendum states are constitutional amendments submitted under a compulsory referendum rather than laws submitted under an optional referendum.

(b) The extent of legislative initiative as compared with popular initiative. Referring again to the table on page 103, it will be seen that of the seven hundred and seventeen measures submitted to a popular vote in the initiative and referendum states, one hundred and sixty were constitutional amendments proposed by popular initiative and one hundred and sixty were laws proposed by popular initiative. That is, out of the seven hundred and seventeeen measures submitted to popular vote three hundred and twenty were initiated by popular petition. That is, about three out of seven of the measures submitted to a popular vote were actually initiated by a popular petition.

Of the seven hundred and seventeeen measures submitted to a popular vote, however, ninety-four laws were referred on the basis of popular petition so that they may properly be added to the total of measures upon which popular petition forced a vote. Adding the ninety-four legislative measures referred on the basis of popular petition to the three hundred and twenty measures initiated by popular petition, we have a total of four hundred and fourteen measures whose submission to a vote was forced by a popular petition. That is, substantially four out of every seven measures were submitted as a result of popular action forcing submision.

The remaining three hundred and three measures were submitted as a result of legislative action, and thirty-four of these three hundred and three were referred to the people by the legislatures as a result of legislative action without waiting for any specific popular demand by petition.

(c) In view of the fact that the compulsory referendum upon constitutional amendments has played such a large part in forcing a large number of popular votes upon measures, it may be well to sum

up the relationship between the optional and the compulsory referendum in the states having the initiative and the referendum. It has already been suggested that of the seven hundred and seventeen measures submitted in these states four hundred and twenty-nine related to constitutional amendments, as to which if the measures were to be adopted a popular vote was compulsory. It is of course impossible to know to what extent under a different machinery the one hundred and sixty initiated amendments would have been handled in another way, but it is possible to say that the present organization of state constitutions made necessary the submission of four hundred and twenty-nine compulsory referenda out of seven hundred and seventeen submissions.

From the standpoint of the voters the thirty-four measures referred by legislatures to a popular vote are compulsory referenda, although of course some of them were optional in the sense that the legislature did not have to submit them in order to bring the measures into effect.

(d) The extent of the popular vote upon measures submitted to the people: It will be of value to review briefly the extent of the popular vote in the states here being dealt with. In Arizona the popular vote upon measures in the elections from 1912 to 1916 presented a fair proportion of the total number of those voting in the general elections; although in 1918 there was a distinct reduction in the popular vote upon measures, and upon several measures the popular vote fell below 50 per cent. In Oregon, where the greatest number of measures have been submitted, the proportion of vote upon measures to the total vote at elections has on the whole been large. For the years 1904 to 1919 the average proportion upon all measures was well over 70 per cent, although there was a slump in popular voting upon measures in 1918. In discussing the proportion of votes in Oregon it should be mentioned that a number of measures have been submitted at special elections. In these cases it is of course improper to compare the total vote with the vote upon measures (inasmuch as the two are substantially identical), and such votes have been excluded from the statement given above. The votes upon measures in Arkansas, South Dakota and Washington have compared favorably with those in Oregon, the average in Washington being above 70 per cent of the total vote at the election. In North Dakota the total vote upon meaures has been relatively large as compared with that upon general elections. In Nebraska, Ohio and Oklahoma the popular vote has been a fair one. In Michigan, for the sixteen measures voted upon at five elections between 1914 and 1918, the popular vote has averaged over 85 per cent. of the total vote for candidates at the same elections.

In Missouri the biennial elections of 1910, 1912, 1914, 1916 and 1918 have brought out relatively large votes upon measures, the votes in 1910, 1912 and 1914 averaging well above 70 per cent of the total vote in the elections. In each of the elections of 1910, 1912 and 1914 in Missouri there was one important question which attracted the attention of the voters. In 1910 the liquor question, in 1912 the tax issue, and in 1914 the issue of woman's suffrage all brought out strong popular votes; and in each of these elections there was an effort on the

part of large interests in the community to have the voters vote "no" upon all measures so that they would be surer of voting "no" upon measures as to which there was the greatest controversy. On this account all measures submitted in Missouri in 1910, 1912 and 1914 were rejected; eleven measures were rejected in 1910, nine in 1912 and fifteen in 1914.

In California popular voting has been uneven. In some elections a good proportion of the electors have voted upon measures, but there are a number of cases in which the total popular vote upon measures falls to 50 per cent or less of the total popular vote in the elections.

Colorado is the one state which makes the worst showing as to popular votes. At the Colorado election in 1912, thirty-two measures were submitted to a popular vote. Of these measures but one received a vote of over 70 per cent of that cast in the general election. Only four measures (including that which received over 70 per cent) received over 50 per cent of the total vote at the election, and twelve had either 30 per cent or less of the total vote cast at the election. Of the sixteen measures submitted in 1914, one had 95 per cent, three above 60 per cent, six a popular vote exceeding 50 per cent of the total vote at the election, (prohibition having 95 per cent), and all of the other measures had cast upon them 50 per cent or less of the total vote at the election. Upon the eight measures submitted in 1916 a better showing was made, two measures having cast upon them 81 per cent of the total vote, and four others having a total vote exceeding 50 per cent. Of the five measures submitted in 1918 almost an equally good showing was made.

Special attention should be called to the results in states which require for the adoption of measures more than a majority of those voting upon the question. Arkansas, Mississippi, Oklahoma and Nebraska require a majority of all votes at the election to adopt constitutional amendments proposed by the legislature, and in these states, as well as in others (such as Illinois) having a similar requirement, it is difficult to obtain such a vote, although Nebaska has in some cases adopted amendments under a plan which obtains party endorsements. and then counts all straight party votes for the measure. Oklahoma requires a majority of all votes at the election to adopt initiated laws or constitutional amendments, and a number of measures have failed because of this requirement; in one case a special election was used for submissions, apparently because a majority at such an election would be substantially equivalent to a majority of those voting on the question.

New Mexico requires a negative vote equal to forty per cent at the election to reject a legislative act on the referendum, but no votes yet have indicated how this provision will work. In Nebraska initiated amendments and laws must receive an affirmative vote equal to 35 per cent of the total vote at the election, and though the Nebraska provision has not been thoroughly tested, the 35 per cent requirement appears to have made no difficulty.

The percentages of affirmative votes on pages 71 and 72 of this pamphlet are of interest in this connection. Upon state-wide public policy questions it will be noted that, while the affirmative votes in only

two cases constitute a majority of all votes cast at the election, several of them run close to 50 per cent, and had they involved real issues, all of them would have resulted in popular approvals, under a rule requiring an affirmative vote equal to 40 per cent of the total vote at the election. Upon these questions it should be borne in mind that the questions are phrased in general language so as to promote affirmative voting, and that the negative vote would probably have been larger except for the fact that it was known that the vote would not be decisive of the isues involved.

Of the state-wide measures appearing on pages 72 and 73 it will be noted that few measures approved by a majority voting thereon would have been rejected had such majority been required to equal 35 or 40 per cent of the total vote at the election. This applies to all state-wide questions submitted, including bank laws, with the exception of the measures submitted during the years from 1892 to 1898. During this period questions appeared upon the general ballot, in such a position that they could not easily be voted upon, but the separate or "little ballot" provided by statute in 1899 removed this difficulty.

The table of popular votes in Chicago since 1905, printed on pages 74 to 79 of this pamphlet, indicates about the same results as does the tables of state-wide votes. Of the questions which received a majority of the votes cast thereon, forty-two had an affirmative vote of more than 40 per cent of the total vote at the election; thirteen between 35 and 40 per cent; and eight less than 35 per cent of such

vote.

Proportion of measures adopted: Little conclusion can be drawn from the proportion between the number of measures adopted and rejected by popular vote. From the table on page 102 it will be seen however that the people have not been inclined to approve all of the measures submitted. The elections of 1910, 1912, 1914 and 1916 show a distinct conservatism upon the part of the electorate.

A table is presented below indicating the total adoptions, classified by methods of submisson, in the states having the initiative and referendum or referendum only. Laws referred on petition are normally referred because some strong opposition justifying the opponents of the legislation in going to the trouble or expense of preparing a petition. The real test of the referendum in this case, therefore, so far as the purpose to be accomplished is concerned, is the proportion of rejections rather than that of adoptions. Tested in this manner the referendum on petition has accomplished its purpose most effectively, having rejected 57.45 per cent of the measures so submitted, whereas the initiative has obtained the adoption of less than 40 per cent of its proposals.

Laws referred by the legislature have the highest proportion of adoptions, and this was probably to be expected in view of the fact that in these cases the legislature was either shifting its own responsibility to the people or was submitting to them a measure upon which