In anticipation of the conference here next month on the Wickersham Commission, I’ve been reviewing the thirteen voluminous reports the Commission issued in 1931 on various aspects of the criminal-justice system. One that holds some interesting surprises is the “Progress Report on the Study of the Federal Courts.” The heart of this report is a fascinating, detailed statistical analysis of the criminal cases in the District of Connecticut for fiscal years 1928-1930.
One thing that strikes me as remarkable is the almost complete absence of trials — the system was dominated then, as now, by guilty pleas. Old-timers today will sometimes tell you about a golden age of trials in the federal system in the 1970′s. In that decade, guilty plea rates hovered between 77% and 82%. After 1981, the rate climbed steadily, reaching more than 96% of adjudicated cases in 2009. But this, apparently, is not a new phenomenon. Among 740 criminal cases filed in the District of Connecticut between 1928 and 1931, only nine went to trial. That’s right, only nine trials in three years, or 1.5 criminal trials per judge per year. (Eight of these trials, by the way, took less than one full day to try.) The guilty plea rate in adjudicated cases was over 98%.
After doing some digging for national data, I discovered that the guilty plea rate rose steadily between 1916 and 1933, reaching a peak of 91%. (See Ron Wright’s helpful data compilation here.) So, Connecticut seems not to have been terribly atypical.
The Connecticut data are, in fact, quite reminiscent of a modern“fast-track” plea-bargaining system.
The fast-track similarities don’t end with the high rate of guilty pleas. Out of the 740 cases, there were only 46 motions to quash or to suppress evidence. And a majority of the cases finished the day the indictment or information was filed, and 85% in less than two months. The indications are of a system that essentially dispenses with formal judicial process — much as we see today, especially in the Mexican border districts, in which fast-track procedures were first developed in order to handle crushing immigration caseloads.
Of course, federal courts in the 1920′s faced their own crushing caseload problem: Prohibition. In 1919, when the National Prohibition (Volstead) Act was adopted, fewer than 21,000 defendants were convicted of federal crimes. By 1932, the number had risen to more than 82,000. This is only a little less than the 87,000 convictions in federal court in 2009 — a shocking fact, when one considers that the U.S. population has more than doubled in the interim.
The conventional story about federalization of criminal law tends to look to the late 1960′s or early 1970′s as a pivotal period, with the federal government both assuming a much higher level of responsibility for basic street crime and greatly expanding the reach of criminally enforceable business regulations. But it is clear that Prohibition was no less a time of extraordinary federalization than our contemporary era. (After the repeal of Prohibition, by the way, annual federal convictions immediately dropped below 42,000 and stayed there — often even below 30,000 — until 1987.)
It seems likely that the extraordinary increase in federal caseloads in the 1920′s contributed to the lack of formal process indicated in the Connecticut data. Circumstances forced the lawyers and judges to figure out how to handle cases in a highly efficient manner, and somehow they managed to do so. (Note, too, that national caseloads and guilty plea rates began to subside in tandem when the Volstead Act was repealed, lending further support to the apparent linkage between the two.)
The Connecticut data reported by the Wickersham Commission, which are purely numerical, do not make clear what greased the wheels of justice. One possibility, as the report suggests, is that prosecutors and/or judges were giving defendants such huge breaks for prompt guilty pleas that defendants could not say no (see, e.g., today’s fast-track inducements). It would be interesting to know, then, how sentencing patterns varied between plea cases and trial cases. However, as the report also notes, there were so few trials in Connecticut between 1928 and 1930 that no conclusions can be drawn from the outcomes in those cases.
In any event, I have thus far tried to highlight some unexpected similarities between then and now. There are, however, also some profound differences. Case composition is one. Today, the federal docket is dominated by drugs, immigration, fraud, and guns, which account for more than 86% of federal felony cases. By contrast, 78% of the cases in the Connecticut study were Volstead Act violations. Yes, there were drug cases in the 1920′s, but they amounted to only 6% of the Connecticut docket. Other crime categories (postal offenses, counterfeiting, etc.) barely registered at all in the data. There were no immigration or gun prosecutions at all.
Sentencing, too, gives us a dramatic contrast between the eras. A fine and suspended sentence was the single most common disposition in the Connecticut study, amounting to nearly 40% of the sentences imposed. Less than 22% of the sentences involved imprisonment. Among the imprisonment cases, the most common sentence length was less than one month. In all, more than 70% of the sentences of imprisonment were for six months are less. The single longest sentence was for 60 months.
In 2009, by contrast, 78% of federal sentences were to prison, and the average term was 56 months. This is an extraordinary turnaround. The ratio of prison to nonprison sentences has been almost exactly reversed (78% non-prison in the old days, 78% prison today), and what is now the average sentence length was only a shade below the very longest in the Connecticut study.
To be sure, the mild sentences in Connecticut must to some considerable extent reflect the peculiarities of the caseload. The crushing number of Volstead Act cases likely led to quite generous sentencing benefits for the overwhelming majority of defendants who pled guilty. Additionally, public support for the Act was not strong — repeal was not far in the future — which may have also contributed to lenience at sentencing. Finally, it is striking that a great majority of the liquor cases were for simple possession.
Yet, even outside the Volstead Act cases, the sentences were hardly draconian. The longest drug sentence, for instance, was 18 months. Average sentence today for drug felonies? 82 months.
Cross posted at Life Sentences.