A new international analysis of countries that hold high proportions of students back to repeat a grade versus those that do not suggests grade repetition is a bad idea. Countries like the United States that have relatively high rates of students repeating a grade do relatively poorly on the tests, while those that have low retention rates do much better. 
[1]
Sarah Sparks at Education Week summarizes [2] the report, from the Organization for Economic Cooperation and Development [3]:
“In comparing the results of the Program for International Student Assessment in 65 member and partner countries, OECD researchers found that differences among countries’ grade-retention trends could explain as much as 15 percent of the difference among their average scores on the 2009 PISA.”
A chicken-and-egg question arises: Is it just the case that students are already higher achievers in the countries with low retention rates and high PISA test scores, and thus don’t need to repeat a grade as often as students in other countries? Perhaps. But Sparks notes that Finland and Korea, which both do well, have banned grade retention (so even if a student did fail a grade, repeating wouldn’t be an option).
Here in the United States, the debate over whether to promote struggling students or hold them back is still raging in many places, and we have written [4]about the high rates of grade retention at many charter schools here at Hechinger. Counterintuitive as it may seem, the evidence – which now includes this report – has largely suggested that social promotion is better for students in the long run than holding them back. Will this report settle the question though? Probably not.