Note: The Border Security, Economic Opportunity, and Immigration Modernization Act was passed by the Senate on June 27 and has moved on to the House.

Pundits are saying the U.S. Congress is about to enact comprehensive immigration reform – that is, legislation combining enhanced enforcement with a path to citizenship for about eleven million undocumented migrants currently living in the United States. Momentum has built since the November 2012 elections put the voting clout of America’s diverse and growing minority groups on full display, and a bipartisan “Gang of Eight” Senators has put forward a bill primed for full Congressional debate. But comprehensive legislation has repeatedly failed before. Will it be different this time?

Although there are no crystal balls, recent history provides sufficient information to make informed predictions. I use some 16,000 earlier Congressional votes on immigration issues to estimate the number of “yes” and “no” votes likely to be cast this time by 535 members of Congress. My analysis suggests that even though a filibuster-proof margin of over 60 votes is well within reach in the Senate, the road to comprehensive reform legislation is much more difficult in the House – and will depend on some legislators changing course.

Using Past Congressional Votes to Predict Future Choices

Here is how my political science model was put together. Taking all immigration-related votes in Congress since the last major push for comprehensive reform in 2005, I sorted out the factors that explain the yes and no votes cast by legislators. Then I use the results to predict how current members of Congress are likely to vote on pending reforms.

• The first step was to create a model that explained as much of the past votes as possible. Previous research pointed to important factors about legislators and their districts – such as partisanship, economic factors, geography, the proportion of immigrants in a district or state, and the weight of particular groups such as the Hispanic/Latino and Asian populations.
• My statistical model considered all votes for all of the legislators, but how well would it predict individual legislators’ choices? To find out, I tested the model against actual voting records. I took all current legislators that were also in office back in 2005, identified key votes that provide strong tests, and then checked to make sure that the model accurately predicts the voting record for each legislator. The results were reassuring. In all, the model provided accurate predictions for 94% of House Representatives and just under 90% of Senators.

Now that the model is available, there are good reasons to forge ahead. Although the 2013 push for immigration reform has its own distinct features, enough features of the current drama echo the past to make predictions worth attempting. Nearly 200 legislators currently in the House of Representatives were also in office during the last round of contentious immigration reform debates in 2005 and 2006, and half of the Senators were there, too. In addition, many key legislative provisions today closely resemble those included in earlier rounds – such as the agreement between organized labor and business to cap guest worker visas at 200,000 per year.

Prospects Look Good in the Senate

In the 2013 Senate, my model projects more than enough votes to get beyond 60 and avoid a filibuster by opponents of reform legislation. According to the detailed data, there are currently 52 solid “yes” votes for the kind of immigration reform pending in the Senate, plus an additional 19 Senators who lean towards voting in favor. The 52 highly likely yes votes are spread across 31 states, and they include eight Republicans as well as Democrats from states like California and New York whom everyone would expect to support reform. As for the nineteen Senators who lean in favor of reform, they include ten Democrats and nine Republicans from 15 states. Opponents of immigration reform are mostly Republicans from the South and inner West.

Is it unrealistic to expect such a large margin of “yes” votes in the Senate? This total is reasonable if we add together Democrats who have supported comprehensive immigration reform in the past, Democrats without a lot of past votes to go on who currently represent diverse states, and Republicans who have supported comprehensive reform in the past.

A final set of likely favorable votes comes from a more debatable group – from Republicans who may not have supported immigration reform in the past, but who represent increasingly ethnically diverse states. Maybe some of these Republicans will not, in the end, favor reform, reducing the total in favor. All in all, my range of prediction is 67 to 71 Senators in favor of reform. But the total could exceed my range if a bandwagon effect sets in, and it could fall short if set-backs develop for reform supporters during Senate debates.

At Best a Cliffhanger in the House

The far tougher hurdle for immigration reform lies in the Republican-controlled House of Representatives. According to my model, between 183 and 203 Representatives are likely to vote yes on comprehensive immigration reform – short of the 218 vote majority mark. However, there is a further step to my analysis. I can also pinpoint the members of Congress most likely to break from positions predicted by past choices.

• Five Democrats with past records suggesting opposition to reform could conceivably switch – under pressure from the vast majority of their own party.
• Because most U.S. voters say they favor reform, potential converts might include House members predicted to vote “no” who face tough re-election contests in 2014. Thirty-three legislators spread across 21 states fit this category, all but four of them Republicans.
• Additional House members whose past records suggest “no” votes come from districts where the numbers of Hispanics/Latinos and Asians who will reach age 18 and be eligible to vote in 2014 eclipse the margins by which these House members won their seats in November 2012.

The bottom line is less exuberant that current pundit predictions. Clearly, it will not be easy for comprehensive immigration reform to pass both houses of Congress in 2013. To be sure, there are several different ways in which a majority could be assembled in the House. But every one of these requires some legislators to break from likely stands. Advocates on both sides of the debate will know the stakes and jump fully into the struggle over the last few critical votes.