Conclusion: Should amendments be introduced and voted on by the full Senate e.g., the Cruz and Sessions amendments that make the initial process of becoming legal contingent on border triggers being met (think triggers first, legalization, and then citizenship instead of legalization first, triggers, and then citizenship), they are likely to fail.
Earlier this week members of the Senate Judiciary Committee (SJC) filed 300 amendments (nearly two-thirds filed by Republicans) to the comprehensive immigration reform bill (S.744) unveiled last month. Today, the SJC will begin debate over these amendments. While we are still a long ways away from the fireworks that are sure to come once the bill is introduced in the full Senate, these amendments provide a preview of what lies ahead. Note: controversial amendments rejected by the SJC are likely to appear again (sometimes in modified forms) once the bill hits the full Senate.
Border security and, specifically, the border triggers that need to be met before legalization can occur, are sure to figure prominently in the debate. Several amendments filed yesterday speak directly to this issue. One can think of the contours of this debate as constituting two main parts. First, what should the border triggers be? Second, where does legalization and a path to citizenship fit in the context of these triggers?
One of the foremost goals of border security – as currently written in the Senate gang of 8’s bill – is an “effectiveness rate” of 90 percent, which roughly translates to mean that 9 out of every 10 illegal entry attempts are detected and deterred at each of the high-risk areas along the Southern border. Based on the amendments that were filed, Senators Cruz (R-TX), Sessions (R-AL), Cornyn (R-TX), and Grassley (R-IA) clearly think that the border triggers need to be stricter. Whereas amendment 1 filed by Senator Grassley makes the simplest point that “effective control” should be required across all parts of the border and not just the high-risk sectors, the other amendments push further by invoking the language of “operational control.”
Amendment 1 filed by Senator Cruz states that the border triggers in the bill should include operational control of “100 percent of the international border between the United States and Mexico” (as well as a tripling of Border Patrol agents and a quadrupling of equipment and technologies). Amendment 11 filed by Senator Sessions similarly seeks to shift the language of border security to operational control. In both cases, the amendments refer to the definition of operational control given in the Secure Fence Act of 2006. Not just a matter of semantics, this seemingly benign shift in language has tremendous implications.
Defining Operational Control
Section 2(b) of the Secure Fence Act of 2006 defines operational control as the “prevention of all unlawful entries into the United States.” The prevention of all unlawful entries across land borders that span nearly 2,000 miles in the south and over 5,000 miles in the north is something that many acknowledge is an unrealistic standard. Even U.S. Customs and Border Protection (CBP) does not use the language of operational control in its own performance and accountability reports. Nevertheless, both the Cruz and Sessions amendments adopt the language of operational control, which sends a clear signal that, for them, the effectiveness rate defined in the gang of 8’s bill is not enough.
Blending Operational and Effective Control… Then Adding Situational Awareness
Amendment 1 filed by Senator Cornyn takes the gang of 8’s concept of effective control and weaves it into a new definition of operational control. Specifically, the Cornyn amendment defines operational control to mean that “within each and every sector of the Southern border, a condition exists in which there is an effectiveness rate, informed by situational awareness, of not lower than 90 percent.” Here, the amendment adds the concept of “situational awareness” to the debate. The amendment defines situational awareness as “the capability to conduct continuous, manned or unmanned, monitoring, sensing, or surveillance of each and every one-mile segment of the Southern border or its immediate vicinity.” To say that situational awareness means 24/7 surveillance of the entire border is no caricature. In fact, the amendment goes on to specify the use of funding to “operate unarmed unmanned aerial vehicles along the Southern border for 24 hours per day and 7 days per week.” Altogether, the Cornyn amendment defines the primary border security goal of the bill as operational control and situational awareness of the entire Southern border (and further specifies that this should be accomplished within a 5-year period).
Operational Control Before Legalization
The gang of 8’s bill lays out a process wherein undocumented immigrants can become legal via registered provisional immigrant (PRI) status, but can only take the steps towards becoming a citizen after certain border triggers are met. Think legal status, triggers, and then citizenship. However, some of the amendments discussed here would change this. Think triggers, legal status, and then citizenship. In other words, these amendments would keep undocumented immigrants undocumented until the border triggers are met. Amendment 4 filed by Senator Grassley states that individuals can only be granted PRI status after effective control of the Southern border has been maintained for 6 months. The Cruz amendment states that its border security requirements must be met “before the Secretary of the Homeland Security may process applications for registered provisional immigrant status.” And lastly, the Sessions amendment states that operational control “triggers the initiation of processing for provisional immigrant status.” To his credit, the Cornyn amendment allows PRI status to come before the triggers are met.
Are These Amendments Likely To Pass?
Altogether, these amendments have the potential to fundamentally alter the immigration reform bill as we currently know it. As Senator Marco Rubio (R-FL) has stated repeatedly for days now, the gang of 8’s bill may need tighter border security measures in order to convince those on the fence – particularly looking forward to the House – to support it. But more to the point for our purposes here, the key question is whether any of these amendments, even if they fail in committee, have a chance to pass when the full Senate takes up debate?
This analysis looks specifically at the likelihood that an amendment that seeks to make legalization contingent on border triggers – for example, the Grassley, Cruz, and Sessions amendments – will pass. While each of these amendments are distinct, one common thread is that each specifies a set of border security measures and then makes legalization contingent upon them.
The Isakson Amendment
In 2006, there was an amendment that was introduced and voted on that did this exact same thing. The Isakson amendment (no. 3961) introduced a “border security certification” section to the immigration reform bill. The amendment stated that until the border security measures identified in the bill have been completed and were fully operational, “The Secretary may not implement any program […] which grants legal status to any individual, or adjusts the current status of any individual, who enters or entered the United States in violation of Federal law.” This amendment was ultimately defeated (40 yays and 55 nays). However, the vote was not neatly predictable based on party lines, as some Democrats joined Republicans in voting for the amendment while some Republicans joined Democrats in voting against it.
In analyzing the vote on the Isakson amendment, we can gain some sense of whether there are enough votes in the Senate to change the current link between border security and legalization from legal status, triggers, and then citizenship to the more pernicious triggers, legal status, and then citizenship. I note here that while this analysis provides insight to the border trigger aspect of the debate, it does not tell us about the other border security amendments that were introduced (e.g., 5,000 more CBP agents). This is planned for a separate analysis.
Just like in 2006, should an amendment be introduced that makes the initial process of becoming legal (not citizenship) contingent on border triggers being met, it is likely to fail. The results indicate 46 yay votes and 54 nay votes (assuming that all Senators vote). The yay votes are broken down into only 15 “solid yes” votes with 31 “lean yes” votes. The nay votes are broken down into 39 “solid no” votes and 15 “lean no” votes. The yay votes include 5 Democrats and the nay votes include 4 Republicans. A full list can be found here.
To be sure, the Isakson amendment is most analogous to the Cruz and Sessions amendments, as they attach tighter border measures to legalization. Given the Cornyn amendment tightens the border security triggers, but maintains the logic of legal status, triggers, and then citizenship, it is likely to garner more support should the full Senate vote on it. It should also be said that, as Senator Grassley proposed two separate amendments on this issue – one dealing with what the triggers should be and the other dealing with how legalization and a path to citizenship fits in the context of the triggers – it may very well be the case that one passes (requiring effective control across the entire border) while the other does not (makes legalization contingent on 6 months of effective control).
Step 1 of this analysis models the vote on the Isakson amendment to the Comprehensive Immigration Reform Act of 2006. Here, I estimate multiple models to identify the best model fit. I then construct categories of likelihood, wherein each predicted probability is characterized as a “solid yes” vote, “leans yes,” “leans no,” and “solid no” vote. These categories take into account the confidence intervals around each predicted probability. Figure 1 provides a graphical illustration.
Step 2 critically evaluates the predicted probabilities obtained in step 1 by analyzing the actual voting record on the Isakson amendment for each current Senator who casted a vote. In all, 35 out of 42 Senators were correctly classified, which represents a match rate of 83.3%.
Step 3 checks the analysis in step 1 against the full history of voting on immigration-related bills in the Senate from 2006 to present as an additional check. The determinants of past votes on immigration policy are analyzed using a logistic regression model: Pr (Voteit = 1|Xit) = P (β0 + β1Xit + εit) where Xit represents a vector of explanatory variables, including key demographics characteristics, and εit represents the error term. In the Senate, there are nearly 6,000 observations to analyze. Predicted probabilities are then obtained from m = 1,000 simulations for each current member of the Senate using the estimated models.
Step 4 synthesizes all of the data – actual voting records and predicted probabilities – and assigns each Senator to one of the four categories of likelihood described above.
Tom K. Wong, Ph.D., is an assistant professor of political science at UC San Diego. He is an expert on immigration politics and policy. He is a research associate at CCIS and beginning in fall 2013 he will be Director of the International Migration Studies Program at UC San Diego. Special thanks to Hillary Kosnac, Cat Benson, Brana Vlasic, and Cameron Kaveh for excellent research assistance.