HAMILTON, Circuit Judge.
Defendant James Barta has appealed his conviction for conspiracy to commit bribery. He was charged and convicted based on an undercover government sting operation that evolved into an agreement among Barta and his co-defendants to bribe a fictional county official in California to obtain a government contract. We reverse because Barta was entrapped as a matter of law. We ordered his release from prison shortly after hearing argument in this case. This opinion provides the detailed explanation for that decision.
“Entrapment is a defense to criminal liability when the defendant was not predisposed to commit the charged crime before the intervention of the government’s agents and the government’s conduct induced him to commit it.”
United States v. Mayfield,
771 F.3d 417, 420 (7th Cir.2014) (en banc). There are two elements to the defense of entrapment: the defendant’s lack of predisposition and the government’s inducement. Here the government made the unusual but appropriate decision to concede that Barta was not predisposed to commit the charged crime. Predisposition is the “principal element in the defense of entrapment.”
United States v. Russell,
411 U.S. 423, 433, 93 S.Ct. 1637, 36 L.Ed.2d 366 (1973). To overcome Barta’s entrapment defense, therefore, the government was required to “prove beyond a reasonable doubt ... that there was no government inducement” in this case.
Mayfield,
771 F.3d at 443. The government failed to do so.
Inducement “means government solicitation of the crime
plus
some other government conduct that creates a risk that a person who would not commit the crime if left to his own devices will do so in response to the government’s efforts.”
Id.
at 434-35. Undisputed evidence shows that several “plus factors” signaling inducement were present in this case. In the course of its undercover operation the government employed “repeated attempts at persuasion.”
Id.
at 435. It employed both “fraudulent representations” and “promises of reward beyond that inherent in the customary execution of the crime.”
Id.
It also employed “pleas based on need, sympathy, or friendship.”
Id.
The cumulative effect of these tactics directed at Barta amounted to inducement. Because Barta was induced and the government concedes that he was not predisposed, it follows that he was entrapped as ,a matter of law. We do not reach the other issues he raises on appeal.
I.
Factual Background
Barta became a target of an FBI sting operation through his friendship with Gustavo Buenrostro. Buenrostro was in turn an associate of Ambrosio Medrano, and both are Barta’s co-defendants in this conspiracy case. The story begins, though, before Buenrostro and later Barta came on the scene. Medrano had dealings with a man purporting to be a health care consultant named George Castro. Castro told Medrano he could, in return for payment of a bribe to a corrupt county official, obtain contract approval from Los Angeles County for the purchase of medical bandages by its hospital system. In reality the man purporting to be Castro was an undercover FBI agent and there was no corrupt official. But Medrano did not know this, of course, so he approached Castro about making another deal.
At that point, in late 2011, Medrano recruited Buenrostro to join him for this
second deal with Castro. For several months, Castro communicated only with Medrano and Buenrostro. The government had no contact with Barta before March 21, 2012, when Barta’s co-defendants first introduced him to Castro. Though this FBI sting stretched out over many months, Barta spoke with Castro four times — in three meetings and one telephone call. These conversations occurred on March 21, May 9, June 12, and June 22. The details of these conversations and the events surrounding each conversation set the stage for the legal analysis below in Section II.
A.
March 21 Meeting
The first thing Barta said to Castro at the March 21 meeting, after the two were introduced to each other, was “I’m not trying to sell you anything.” Barta told Castro that he was at the meeting because of Buenrostro: “Gus asked me to come and tell you what we do.” The “we” in that statement refers to Sav-Rx, which Barta described to Castro as a prescription benefit management business. Barta is a founder and owner of Sav-Rx, though he turned over much of the management of the company to his daughter in 2007. He told Castro that Sav-Rx provided prescription benefit management services primarily for unions, but also for the Cook County hospital system in Chicago.
Castro responded that while his company was not currently involved with pharmaceuticals, he could serve as a “broker” between companies and Los Angeles County because he knew “an individual in the county system.” At different points in the meeting Castro described the payment he was seeking in return for the promised contract approval as a “finder’s fee,” “good faith money,” and a “commission.” To these various descriptions Barta replied that he understood what Castro was saying, and Barta closed the meeting by saying “I hope we can do some business.” Barta spent just twelve minutes at this meeting.
A bit more background about the sting operation provides the context for the March 21 meeting. Medrano was introduced to Castro by a man named Michael DiFoggio, who was a cooperating witness working with the FBI as part of this sting operation. Before the March 21 meeting, DiFoggio repeatedly asked Medrano to send a written proposal for a deal between Sav-Rx and Los Angeles County in advance of the meeting. Medrano did not do so, and Barta himself never provided such a proposal, though he did give Castro some generic Sav-Rx brochures at their first meeting on March 21. DiFoggio also repeatedly asked Medrano for a cash payment from Barta to Castro. DiFoggio asked for this payment to be made at the March 21 meeting. It was not. DiFoggio then asked that the payment be made on May 9, the next time that Barta and Castro met. Barta made no payment at that meeting either.
Nothing in the previous paragraph should be taken to suggest that before the March 21 meeting DiFoggio directly asked Barta for anything. At that time DiFog-gio was dealing only with Medrano and Buenrostro. It was they who told DiFog-gio that Barta would be willing to provide a proposal and payment. Barta had no contact with the government (Castro or DiFoggio) until the March 21 meeting.
B.
May 9 Meeting
The May 9 meeting was Barta’s longest meeting with Castro. Medrano and Buen-rostro began the meeting by explaining how they could form a company to partner with Sav-Rx as a “minority participant” in a contract with Los Angeles County. Castro continued the conversation by saying he spoke with “the guy that would approve” the contract, and this person would
ensure approval of the contract by tailoring the request for proposal from the county to favor Sav-Rx.
A quiet Barta had not said much in the meeting when Buenrostro asked him, “What are your views on all this stuff, Jim, or are you just soaking it in?” Barta responded that Sav-Rx had succeeded in Cook County by cutting the amount of time patients waited for their prescriptions and the amount of money hospitals paid to dispense prescriptions. He went on to say that if Los Angeles County had “a problem,” as Cook County had earlier, then a deal would make sense. But if Los Ange-les County was already “efficiently dispensing” its medications, then there would be no problem for Sav-Rx to solve. Barta also described how adding a mail-order option had created greater efficiencies in Cook County. Near the end of the meeting Barta explained to Buenrostro how Buenrostro could research the Los Ange-les County hospital system to determine whether it would benefit from Sav-Rx’s services. And Barta told Castro that if Los Angeles County had a problem for Sav-Rx to solve, “I’m in 100%.”
After the May 9 meeting the FBI aggressively tried to close the fake deal — and the real operation underlying it. Here are excerpts from emails that Castro sent to Barta following that meeting, none of which drew a reply from Barta. “I spoke to my friend and he is willing to move forward on our end. We await your response if you would like to proceed.” That email was sent May 16. “My friend would like to get this matter settled and I told him that by next week we would know whether or not we are moving forward.” That email was sent May 30. “We are ready to begin on our end. I will be in Chicago until Friday. I can extend my visit through as late as next Thursday if you want to meet in person to ... take care of the final detail. Let me know how you would like to proceed.” That email was sent June 6. Again, Barta replied to none of these emails.
The FBI also sent Barta emails about the Los Angeles County hospital system. Confronted at the May 9 meeting with Barta’s concern about whether the county had a problem for Sav-Rx to solve, the FBI decided to fabricate a problem. At the time the real Los Angeles County hospital system was already offering a mail-order option for receiving prescriptions. It was also using automation at its pharmacies. Buenrostro’s research revealed as much. The FBI responded by falsely claiming that patients were waiting too long for prescriptions and were receiving the wrong prescriptions, along with falsely claiming that the automation technology was inefficient and unreliable.
As the sting operation stretched on, the FBI also sweetened the terms of the tentative deal being discussed. In the May 9 meeting, Castro mentioned a three-year contract to dispense around 1 million prescriptions per year. Later in May, Castro sent emails substantially raising the financial stakes for the fictional deal, saying the number of yearly prescriptions would be up to 2 million in the first year of the contract and 3 million in the remaining years.
Castro next met with Medrano and Buenrostro on May 24. Barta was not at that meeting. Castro complained that Barta had made no payment at the May 9 meeting, but Buenrostro reassured Castro that Barta would commit to the deal in the end. Buenrostro explained that Barta would do so because Barta “wants to help me get back on my feet.”
C.
June 12 Phone Call
Castro sent a number of emails to Barta in the month preceding this June 12 phone call, as discussed above. Barta replied to
none of them. Castro then began calling Barta. His first call on June 7 went unanswered, and he left a voicemail for Barta. The same thing happened again when he called on June 11. Having received no response to his emails or calls, Castro issued an ultimatum to Barta on June 12. He wrote an email to Barta and his co-defendants: “If I don’t hear anything from Jim by the end of the day, I will assume he is no longer interested and my guy will move on. I hope to hear from you.” Bar-ta still did not reply. Yet Castro did not follow through on his ultimatum.
Instead, after sending the email Castro called Barta later that same day. Barta’s assistant may have transferred this call to him by mistake because Barta quickly told Castro that “you just caught me flat footed. I’m in the middle of something else, okay?” In answer to Castro’s question of whether Barta was “ready to move forward” with the deal Barta said, inconclusively, “I think we’re probably ready to move.... Yep.” This conversation lasted less than two minutes.
Hours after speaking with Barta, Castro called Buenrostro. This was one of over fifty contacts that Castro had with Buen-rostro in June, several of them urging Buenrostro to elicit a response from Barta. Castro and Buenrostro set up a meeting for June 19 in Chicago. (Chicago was also the site of the March 21 and May 9 meetings.) But when Castro called Buenrostro on June 18 to confirm the meeting scheduled for the next day, Buenrostro told him that Barta was not going to be there. Buenrostro suggested that they travel to Nebraska, where Barta lived, to conclude the deal. Castro preferred to meet in Chicago, and he left a voicemail for Barta on June 18. He left two more voicemails for Barta on June 19. None of these calls were returned.
D.
June 22 Meeting
Ultimately, Castro traveled to Nebraska on June 22 to meet with Barta. At this final meeting Buenrostro joked with Barta (ironically, in our view of the case) that, “It took a lot to get you here.” But later in the meeting Barta made clear why he was there. He explained: “At this stage in my career I could give a damn or less about ... whatever happens.... But I’d like to see Gus do something. You know Gus, Gus has always been a day late and a dollar short his whole damn life.” That final meeting ended with Barta writing Castro a Sav-Rx check for $6,500 to be used to pay the fictional county official.
Six days later Barta was arrested. Bar-ta was charged with conspiracy to commit bribery under 18 U.S.C. § 371. He was tried before a jury, which was instructed about the entrapment defense. The jury found Barta guilty. After the jury verdict Barta moved for a judgment of acquittal based on his entrapment defense. The district court denied the motion, and Barta has appealed.
II.
Analysis
A.
Entrapment
“In entrapment cases, the government must prove beyond a reasonable doubt that a defendant who raises a color-able defense of entrapment ... has not in fact been entrapped.”
United States v. Theodosopoulos,
48 F.3d 1438, 1444 (7th Cir.1995). Whether the government has met this burden of proof in a given case is generally a question for a jury.
Id.
at 1445. When a jury decides that the government has met its burden, then a court reviewing that decision must examine “all the evidence in the record in the light most favorable to the government,” while still ensuring that “the government presented sufficient evidence to permit a jury to de
termine beyond a reasonable doubt” that the defendant was not entrapped.
Id.
Though this is a favorable standard of review for the government, it does not guarantee that the government will prevail. See, e.g.,
Jacobson v. United States,
503 U.S. 540, 112 S.Ct. 1535, 118 L.Ed.2d 174 (1992) (finding defendant was entrapped as a matter of law);
Sherman v. United States,
356 U.S. 369, 78 S.Ct. 819, 2 L.Ed.2d 848 (1958) (same). The evidence here shows entrapment beyond reasonable dispute.
“Entrapment is a defense to criminal liability when the defendant was not predisposed to commit the charged crime before the intervention of the government’s agents and the government’s conduct induced him to commit it.”
Mayfield,
771 F.3d at 420. The first element of an entrapment defense, predisposition, is the “principal element.”
Russell,
411 U.S. at 433, 93 S.Ct. 1637. In this case the government conceded at trial that Barta was not predisposed to conspire to commit bribery. The focus is thus on the second element, inducement. The government was required to prove beyond a reasonable doubt that it did not induce Barta to conspire with his co-defendants.
Mayfield,
771 F.3d at 443.
In
Mayfield,
a recent en banc decision surveying entrapment law, we explained that inducement “means government solicitation of the crime
plus
some other government conduct that creates a risk that a person who would not commit the crime if left to his own devices will do so in response to the government’s efforts.”
Id.
at 434-35. Several of the plus factors signaling inducement that we identified in
Mayfield
are present in this case. A quick review of the facts makes this clear. The key facts are not disputed. They are taken largely from email records and recorded conversations that were collected by the government as part of its undercover operation.
The FBI frequently emailed and called Barta, with no response from Barta. These were “repeated attempts at persuasion.”
Id.
at 435. The FBI invented false deadlines for Barta to commit to the deal and invented false problems for the Los Angeles County hospital system. These were “fraudulent representations.”
Id.
The FBI significantly sweetened what would have already been an attractive deal to Barta and his co-defendants. Here we have “promises of reward beyond that inherent in the customary execution of the crime.”
Id.
And the FBI pressed Barta— both directly and through Buenrostro — to make a deal that it had reason to believe Barta would be making mainly to benefit his less fortunate friend, Buenrostro. Here we have “pleas based on need, sympathy, or friendship.”
Id.
The presence of all these plus factors shows that the government induced Barta to commit a crime, one that the government concedes he had no predisposition to commit. That is enough to establish entrapment as a matter of law.
The cases the government cites are not to the contrary. The government first takes on the “repeated attempts at persuasion.” It argues that Castro did not contact Barta enough, or have a lengthy enough relationship with him, to induce him to commit a crime. See
United States v. Plowman,
700 F.3d 1052, 1059 (7th Cir.2012) (rejecting entrapment defense where defendant and undercover agent interacted for “a mere five months”);
Theodosopoulos,
48 F.3d at 1447 (rejecting entrapment defense where interaction between defendant and government “merely stretched over the course of three months and nine meetings”). But there is no
per se
rule regarding the number of contacts or
length of relationship it takes to constitute inducement.
Each case, and each entrapment defense, must be judged on its own facts.
Plowman
and
Theodosopoulos
did not adopt mechanical rules. It is not just the number of contacts between Castro and Barta or the length of their relationship-that amounted to inducement here. It was also the frequency of those contacts. And the fact that the emails and calls to Barta received no responses — even when they gave fake ultimatums. And the fact that the contacts occurred alongside other government conduct intended to induce Barta to go forward with the criminal plan: the fabricated problems in Los Angeles County’s handling; of prescriptions; the sweeteners added to an already lucrative deal; and the efforts to appeal to Barta based on his friendship with Buenrostro.
The government argues that these other forms of pressure on Barta did not constitute plus factors signaling inducement. It does so in much the same way that it argues this case presents no “repeated attempts at persuasion,” namely by extrapolating
per se
rules from precedents addressing each type of plus factor in isolation. For example, the government summarizes
United States v. Millet,
510 F.3d 668, 677 (7th Cir.2007), by stating that promises of reward totaling several hundred thousand dollars did not create inducement in that case. That citation to
Millet
is accurate, so far as it goes, but we decline to decide this case by analyzing each plus factor as if it were a rule unto itself. We do so for an important reason that is worth explaining.
There is little to be gained and much to be lost in trying to formulate precise markers for when each plus factor is present or absent. The plus factors are supposed to guide a jury or judge in making a more general determination about inducement. “Multifactor tests are common in our law but they can be cryptic when unattached to a substantive legal standard .... Knowing what factors to look
at
is useless unless one knows what to look
for.” Mayfield,
771 F.3d at 435. Formulating rigid rules for applying each plus factor would risk turning the plus factors into barriers rather than aids to good judgment.
Mayfield
explains what to look for in applying the plus factors indicating inducement: government conduct creating a risk that Barta committed a crime he otherwise would not have committed.
Id.
at 434-35. Determining whether such government conduct occurred — whether any plus factors were present — is a contextual inquiry that cannot be reduced to applying a checklist of rigid rules. Determinations of inducement (and predisposition) must be made by considering all the facts and circumstances of each case. On the facts and circumstances of this case, several plus factors were indisputably present, establishing government inducement that entrapped Barta as a matter of law.
Because Barta points to much more than just government persistence to establish entrapment, we need not consider what degree of persistence alone would suffice to show inducement as a matter of law. But make no mistake, the government was very persistent with Barta. We reject the government’s contention that the “repeated attempts at persuasion” here were not strong evidence of inducement. Between the May 9 and June 22 meetings the FBI sent five emails and placed six calls to Barta. Barta never answered any of these emails or returned any of these calls.
The government offers an explanation for its persistence. It argues that because Barta was a busy and cautious businessman, it made sense that he did not have the time or inclination to communicate his commitment to the conspiracy as often or
as clearly as his co-defendants signaled theirs. This argument is premised on a misunderstanding of why persistence counts as inducement and why inducement is an element of entrapment.
Persistence counts as inducement because we worry that if “additional efforts at persuasion” by the government are required to convince someone to commit a crime,
id.
at 431, then the result will be “the apprehension of an otherwise law-abiding citizen who, if left to his own devices, likely would have never run afoul of the law.”
Jacobson,
503 U.S. at 553-554, 112 S.Ct. 1535. And inducement is an element of entrapment because those who accept an “ordinary opportunity to commit the crime without additional efforts at persuasion” are justifiably punished, whether predisposed or not.
Mayfield,
771 F.3d at 431.
It is thus not relevant to the inducement inquiry whether “additional efforts at persuasion” are required because the target of the sting operation is too busy with other things, too reverent toward the law, too cautious about the possibility of a government sting, or whatever the case may be. The point is that the government is supposed to catch criminals, not create them. The government’s conduct here, including its persistence, posed an impermissible risk that Barta’s criminality was created rather than caught.
B.
Conspiracy
Why did the government go to such great lengths to get Barta finally to make a payment to Castro? The answer can be found in conspiracy law. “To convict a defendant of conspiracy, the government must prove that (1) two or more people agreed to commit an unlawful act, and (2) the defendant knowingly and intentionally joined in the agreement.”
United States v. Johnson,
592 F.3d 749, 754 (7th Cir.2010). A “mere association with conspirators” or “knowledge of a conspiracy,” even when combined with “presence during conspiratorial discussions,” is not sufficient “to convict a man of conspiracy.”
United States v. Garcia,
562 F.2d 411, 414 (7th Cir.1977). The government needed evidence that Barta “joined and participated in the conspiratorial scheme.”
Id.
The government obtained that evidence on June 22 when Barta wrote Castro a check. (That Barta wrote this check from a company account is one indication among many that he was not the cautious and calculating criminal mastermind that the government made him out to be, but we put that aside.) The government went so far because it wanted to obtain such definitive evidence. But it obtained that evidence through entrapment, as explained above. And this may be why the government argues on appeal that Barta joined the conspiracy on or before March 21, before any of the inducement occurred. This argument is flawed because it attempts to uphold a jury verdict on a theory never presented to the jury.
Theories of guilt or liability not argued to the jury are waived on appeal. See
Absher v. Momence Meadows Nursing Center, Inc.,
764 F.3d 699, 711-12 (7th Cir.2014) (collecting cases);
United States v. Mohamed,
759 F.3d 798, 808-09 (7th Cir.2014). At trial, the government argued that Barta joined the conspiracy on May 9 or June 12, and it asked the jury to return a guilty verdict on that basis. In its closing argument at trial, the government acknowledged that Medrano and Buenrostro had formed a conspiracy before Barta allegedly joined, and then asserted that “Barta joins at some point later, whether it’s on the 9th or the very, very latest, that phone call on June 12th.” Trial Tr. 1838. This closing argument— that Barta joined the conspiracy on May 9 or June 12 — cannot be squared with the
government’s new theory of the case on appeal, that Barta joined the conspiracy on or before March 21.
If Barta had in fact joined a conspiracy with Medrano and Buenrostro on or before that first contact with the government on March 21, the entrapment defense would not be available. It was after all Buenros-tro and Medrano who first brought Barta into the discussions with Castro. The government cannot be held responsible for those actions. But the theory of an earlier conspiracy would have to be based on inferences drawn from Barta’s minimal comments at the March 21 meeting, along with statements by the alleged co-conspirators to Castro about Barta’s supposed agreement. That theory would have made for a very thin case against Barta, if it could have stood at all. We need not and do not determine here whether the government might have been able to win its case against Barta on that theory. The problem is that the government did not argue that theory to the jury that convicted Bar-ta.
The government’s theory on appeal is also inconsistent with the testimony of its undercover agent. The agent posing as Castro testified that, as of June 5, Barta “had not confirmed yet” and “was not on board” with the bribery scheme. Tr. 896. These statements undermine the government’s new theory. It is true that “Castro” later in his testimony added that Barta “was always on board with the payment,” though he was not on board with “the actual services or the way that the Sav-Rx company was going to be able to work into the county system” until after June 5. Trial Tr. 998. This later testimony is hard to reconcile with a great volume of evidence here, but the government interprets Castro’s later testimony as indicating that Barta joined the conspiracy long before June 5 and merely continued to work out the details of the conspiracy after that date.
The government appears to be relying on the principle that “a conspiracy may be actionable, even though it is conditional.” See
United States v. Podolsky,
798 F.2d 177, 178 (7th Cir.1986). On this view of the evidence, perhaps Barta might have made a “conditional agreement” to commit bribery.
Id.
But while conditional agreements can be actionable as conspiracies, it is also true that if the condition is a condition not of committing the crime “but of agreeing in the first place, there is no criminal conspiracy unless and until the condition is satisfied.”
Id.
On this other view of the evidence, Barta made at most “an agreement to negotiate an agreement” with his co-defendants, which would not have amounted to a crime.
Id.;
see also
United States v. Melchor-Lopez,
627 F.2d 886, 891 (9th Cir.1980) (“The line between conspiracy and an unexercised opportunity to join a conspiracy may be difficult to draw, but it must be drawn where the existence of an agreement is absent.”).
We doubt that a prosecutor armed only with the evidence that existed at the end of the day on March 21 would actually press this “conditional agreement” theory by pursuing a case against Barta. Prosecutors are busy officials with better cases to pursue. We also doubt that a jury could be convinced beyond a reasonable doubt to convict Barta based on this “conditional agreement” theory. But given how the government handled the case, as opposed to how it might have handled the case, these are questions that we need not ponder.
Barta’s conviction is REVERSED and the case is REMANDED with instructions to enter a judgment of acquittal.