Long v. Social Security Administration

Summarize
Check Treatment

Related Cases

United States Court of Appeals for the Federal Circuit __________________________ DANVERS E. LONG, Petitioner, v. SOCIAL SECURITY ADMINISTRATION, Respondent. __________________________ 2010-3108 __________________________ Petition for review of the Merit Systems Protection Board in case no. CB7521080019-T-1. __________________________ Decided: March 14, 2011 __________________________ CHRISTOPHER C. SHARP, Sharp Law Firm, P.A., of Plantation, Florida, argued for petitioner. ELIZABETH M. HOSFORD, Senior Trial Counsel, Com- mercial Litigation Branch, Civil Division, United States Department of Justice, of Washington, DC, argued for respondent. With her on the brief were TONY WEST, Assistant Attorney General, JEANNE E. DAVIDSON, Direc- tor, and BRIAN M. SIMKIN, Assistant Director. LONG v. SSA 2 LINDA A. STAGNO, Association of Administrative Law Judges, of Brooklyn, New York, argued for amicus curiae. Of counsel was HAROLD J. KRENT, Chicago-Kent College of Law, of Chicago, Illinois __________________________ Before GAJARSA, DYK, and PROST, Circuit Judges. Opinion for the court filed by Circuit Judge PROST. Concurring opinion filed by Circuit Judge DYK. PROST, Circuit Judge. Petitioner Danvers E. Long (“Long”) petitions for re- view of the final decision of the Merit Systems Protection Board (“Board”) finding good cause to remove Long from his position as an administrative law judge (“ALJ”) with the Social Security Administration (“Agency”) based on a charge of conduct unbecoming an ALJ. We affirm. BACKGROUND The Agency employed Long as an ALJ in its Office of Disability Adjudication and Review in Fort Lauderdale, Florida beginning in 2001. At approximately 11:00 p.m. on January 27, 2008, Long was involved in a physical altercation (“January 27, 2008 Incident”) with his domes- tic partner, Lilia Vanessa Castro (“Castro”), who is the mother of his youngest child, Charlize Long Castro. The physical altercation began when Castro returned home, after leaving their child in Long’s care, to discover that Long was asleep in their bedroom and the child was not with him. Eventually, Castro fled the home on foot and continued to the home of Lisa and Donald Feeney (“the Feeneys”), who lived almost a mile away from Long and Castro. After Castro told the Feeneys that Long hit her 3 LONG v. SSA and the child, Lisa Feeney called 911 on Castro’s behalf. Three police officers responded to the Feeneys’ home and questioned Castro; a Spanish-speaking officer took a tape- recorded statement from Castro. In these interviews and Castro’s statement, Castro explained that Long repeat- edly struck and pushed her and that he accidentally struck their child. The police observed, and took digital photographs of, physical injuries on Castro’s face, fore- arm, and thigh as well as a red mark on the child’s face. Long was arrested and, on February 21, 2008, was charged with one count of domestic violence battery and one count of culpable negligence. The prosecutor later entered a nolle prosequi on the criminal charges against Long. The Agency filed a Complaint, on June 30, 2008, and an Amended Complaint, on October 3, 2008, seeking Long’s removal from his position as an ALJ based on one charge of conduct unbecoming an ALJ arising out of the January 27, 2008 Incident. The charge contained two specifications: Specification 1: On or about January 27, 2008, [Long] repeatedly struck, grabbed, and pushed Lilia Vanessa Castro. Specification 2: On or about January 27, 2008, [Long] struck Charlize Long Castro, while Lilia Vanessa Castro was holding her. J.A. 72. At a hearing before Administrative Law Judge Gian- nasi (“ALJ Giannasi”), sitting by designation, ALJ Gian- nasi heard testimony from, inter alia, Long, Castro, Lisa Feeney, Donald Feeney, and two of the three responding police officers. The parties jointly stipulated to the sub- mission of the deposition testimony of Long’s teenage LONG v. SSA 4 children who were home during the January 27, 2008 Incident, Ana Long and Danvers Long, Jr., in lieu of having the two testify at the hearing. Lisa Feeney, Don- ald Feeney, and the two police officers testified that Castro, who was noticeably shaken and had visible physi- cal injuries, told them that Long was physically violent. Long, however, denied striking Castro and the child but admitted that, during the January 27, 2008 Incident, he grabbed Castro on several occasions and may have pushed her. Long claimed that his actions during the incident were taken in defense of himself and his child. Castro, who by the time of the hearing had reconciled with and was again living with Long, denied telling the Feeneys and the police that Long struck her and the child during the January 27, 2008 Incident. She testified that Long only grabbed her to protect himself and did not push or strike her. She further testified that she did not know if Long struck the child. On June 2, 2009, ALJ Giannasi issued an initial deci- sion sustaining the charge against Long. ALJ Giannasi found that, though the details of the January 27, 2008 Incident were not clear, the Agency established that Long used violence against Castro and thus proved by a pre- ponderance of the evidence the “substance” of the first specification of the charge. Soc. Sec. Admin. v. Long, No. CB7521080019-I-1, slip op. at 21-22 (M.S.P.B. June 2, 2009) (“Initial Decision”). ALJ Giannasi, however, found that the Agency failed to prove the second specification. Id. at 22-23. Based on his review of the Douglas factors, ALJ Giannasi found good cause for a forty-five day sus- pension, rather than the requested penalty of removal. Id. at 1, 31-32. The Agency petitioned the Board for review of the ini- tial decision. Soc. Sec. Admin. v. Long, No. 5 LONG v. SSA CB7521080019-I-1, slip op. at 1-2 (M.S.P.B. Jan. 27, 2010) (“Final Decision”). The Board granted the Agency’s petition and issued a final decision on January 27, 2010. Id. It found that the Agency proved both specifications of the charge of conduct unbecoming an ALJ by a prepon- derance of the evidence. Id. at 2, 6, 20. In reaching this conclusion, the Board, in contrast to ALJ Giannasi, found that the hearing testimony of Long and Castro was not credible or reliable. Id. at 6, 18-20. The Board instead relied on the hearing testimony of the Feeneys and the police officers, the police reports, Castro’s sworn state- ment to police the night of the incident, and the statement and deposition testimony of Ana Long. Id. at 6, 8, 14-20. Further, the Board found good cause for disciplinary action against Long and determined that removal was the appropriate penalty. Id. at 20-27. Long timely petitioned for review of the Board’s final decision in this court. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(9). DISCUSSION “Our review of Board decisions is limited. We may only reverse a Board decision if we find the decision to be arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; obtained without procedures required by law; or unsupported by substantial evidence.” Kahn v. Dep’t of Justice, 618 F.3d 1306, 1312 (Fed. Cir. 2010) (citing 5 U.S.C. § 7703(c)). Long raises a number of issues on appeal. First, Long argues that the Board’s finding that the Agency proved both specifications of the charge of conduct unbecoming an ALJ is not supported by substantial evidence. Second, Long contests the Board’s finding of “good cause” for LONG v. SSA 6 disciplinary action against him. Finally, Long objects to the imposed penalty of removal. We address each in turn. I A Long argues that the Board’s decision to sustain the charge of conduct unbecoming an ALJ is not supported by substantial evidence because the Board improperly over- turned ALJ Giannasi’s demeanor-based credibility deter- minations, including his determination that the testimony of several Agency witnesses was not fully credible or reliable as well as his concomitant finding that the Agency witnesses who spoke with Castro the night of the incident were unable to obtain a full understanding of the incident given Castro’s limited ability to speak English. Pet’r’s Br. 19, 22-41; Reply Br. 4-20. In response, the Agency contends that the findings with which Long takes issue on appeal were not demeanor-based determinations and thus the Board was free to substitute its judgment for that of ALJ Giannasi. Resp’t Br. 20-32. The general rule is that the Board is free to re-weigh the evidence and substitute its judgment for that of one of its administrative judges. Leatherbury v. Dep’t of the Army, 524 F.3d 1293, 1304 (Fed. Cir. 2008); Haebe v. Dep’t of Justice, 288 F.3d 1288, 1302 (Fed. Cir. 2002). “An important exception, however, is that the Board is not ‘free to overturn an administrative judge’s demeanor based credibility findings merely because it disagrees with those findings.’” Leatherbury, 524 F.3d at 1304. Specifically, where an administrative judge [is] able to observe the demeanor of a testifying witness and, as a re- 7 LONG v. SSA sult, the administrative judge’s findings [are] ex- plicitly or implicitly based on the demeanor of the witness, the Board may not simply disagree with the AJ’s assessment of credibility . . . unless the [B]oard has articulated sound reasons, based on the record, for its contrary evaluation of the tes- timonial evidence. Id. at 1304-05 (quotation marks omitted). Therefore, “if the [Board]’s reasons for overturning demeanor-based credibility determinations are not sufficiently sound, its decision does not survive substantial evidence review.” Haebe, 288 F.3d at 1301. Accordingly, for each of ALJ Giannasi’s determina- tions that the Board overturned, if the determination was not demeanor-based, there was no restraint on the Board’s ability to reconsider the evidence and reach a different conclusion. If the determination was explicitly or implicitly demeanor-based, however, the Board was required to give “sound reasons, based on the record” for overturning ALJ Giannasi’s conclusion. We need not address into which category the determinations at issue on appeal fall because we conclude that the Board, in its detailed, thorough analysis of the evidence and ALJ Giannasi’s determinations, gave sound reasons for any conclusions that were contrary to those reached by ALJ Giannasi. 1 The Board therefore satisfied the more strin- 1 There is, however, one exception. Specifically, we conclude that the Board erred in overturning ALJ Gian- nasi’s findings regarding Castro’s provocation of Long but this error was harmless. In its analysis of provocation, the Board stated “Ms. Castro struck the respondent in an attempt to get away from him after he pushed and grabbed her when she woke him up.” Final Decision at 26 (emphasis added). As such, the Board appears to be LONG v. SSA 8 gent standard for overturning demeanor-based credibility determinations. For example, Long objects to the Board’s disagree- ment with ALJ Giannasi’s findings regarding the reliabil- ity and credibility of Sergeant Coleman, one of the responding police officers. Pet’r’s Br. 38; Reply Br. 31-32. ALJ Giannasi found that Sergeant Coleman’s testimony was entitled to less weight because certain discrepancies led him to question its accuracy. Initial Decision at 13. First, ALJ Giannasi noted that Sergeant Coleman testi- fied that Castro stated that she left the house after Long prevented her from calling the police, yet this fact is not mentioned in Sergeant Coleman’s incident report and is not “supported by any other witness or any other evi- working on the premise that Long was the first to use physical violence. The Board did not acknowledge that Castro’s sworn statement as well as Castro’s and Long’s hearing testimony state that Castro struck Long before he struck her. J.A. 99-100; J.A. 1024; J.A. 1206-08. Because there is no contradictory evidence in the record, this point is undisputed. Based on this uncontroverted record evidence, ALJ Giannasi properly found that Castro initi- ated the violence. Initial Decision at 15, 21, 31. The Board’s error, however, was harmless because its ultimate conclusion regarding provocation is unaffected by its premise that Long was the first to use physical violence. The Board ultimately determined that provoca- tion is not a mitigating factor “lessening [Long]’s culpabil- ity for his conduct,” because Long “escalated” the situation by pursuing Castro “throughout the house and eventually out into the neighborhood,” rather than choos- ing to avoid “further conflict.” Final Decision at 26. Regardless of whether Castro struck Long first, the Board’s conclusion that provocation is not a mitigating factor, based on Long’s escalation of the violence despite having the opportunity to avoid further conflict, is well- supported by the record evidence and appropriate. 9 LONG v. SSA dence.” Id. at 12-13. Second, ALJ Giannasi explained that he “doubt[ed] Sergeant Coleman’s testimony that Ms. Castro told him that she was hit by a closed fist” because “[s]he did not tell the Feeneys or anyone else that she was hit by a closed fist.” Id. at 13. The Board, however, rejected ALJ Giannasi’s conclusions regarding Sergeant Coleman’s testimony. Final Decision at 16. As to the first alleged discrepancy, the Board pointed to other record evidence referencing that Castro left the house after Long prevented her from calling the police, including Sergeant Coleman’s own probable cause affidavit, Officer Madison’s incident report, and Castro’s sworn statement, thereby belying ALJ Giannasi’s finding that the fact was not “supported by any other witness or any other evidence.” Id.; J.A .83; J.A. 85; J.A. 94-95; see also J.A. 104. With respect to the second alleged discrepancy, the Board cited record evidence showing that Castro had indeed stated to other officers that Long hit her with a closed fist, includ- ing her sworn statement and Officer Madison’s incident report. Final Decision at 16; J.A. 82-83; J.A. 92. Because the Board referenced specific portions of the record estab- lishing that ALJ Giannasi’s findings regarding discrepan- cies in Sergeant Coleman’s testimony were factually inaccurate, the Board undoubtedly provided “sound reasons, based on the record,” for its contrary conclusions. Additionally, Long disputes the Board’s rejection of ALJ Giannasi’s determination that testimony from the witnesses who spoke with Castro the night of the January 27, 2008 Incident was not necessarily accurate or reliable. ALJ Giannasi based this determination on his finding that Castro’s limited ability to speak English prevented the witnesses from obtaining a full understanding of the incident. Initial Decision at 16, Pet’r’s Br. 30-33. Long defends ALJ Giannasi’s determination that the witness testimony was unreliable, emphasizing that the police LONG v. SSA 10 used a Spanish-speaking officer to interview Castro and that she testified in Spanish through an interpreter at the hearing before ALJ Giannasi. According to Long, these facts demonstrate Castro’s limited English abilities. Nevertheless, we conclude that the Board gave “sound reasons” for its disagreement with ALJ Giannasi’s deter- mination. Indeed, as the Board noted, with specific citations to the record, “all witnesses at the hearing who spoke with Ms. Castro on the night of the incident, i.e., Mr. and Mrs. Feeney and two police officers, indicated that they had no trouble understanding her.” Final Decision at 14-15; J.A. 1159-64; J.A. 1170; J.A. 1188; J.A. 1197; J.A. 1241; J.A. 1254-56; J.A. 1281. In addition, the Board explained that though it may have been easier for a Spanish speaker to obtain details of the incident from Castro, there is no evidence that she was unable to com- municate the relevant facts in English. Final Decision at 15. This conclusion is supported by testimony from Sergeant Coleman that he was able to communicate with Castro and directed the Spanish-speaking officer to interview Castro only to allow her to communicate with ease. Id.; J.A. 1254-56. Again, because the Board pointed to specific record evidence that contradicted ALJ Gian- nasi’s determination, we conclude that the Board’s rea- soning was “sufficiently sound.” B Long also argues that the Agency did not prove either specification of the charge of conduct unbecoming an ALJ. Pet’r’s Br. 19, 42-53; Reply Br. 34-37. Long’s argument, however, is based on ALJ Giannasi’s findings regarding the January 27, 2008 Incident, which Long urges the court to adopt. See Pet’r’s Br. 44, 51. Because we have instead upheld the Board’s findings regarding the Janu- 11 LONG v. SSA ary 27, 2008 Incident, we need not address the specifics of Long’s argument in this regard. Further, Long asserts that the Agency bore the bur- den to prove the elements of the criminal charge of mis- demeanor domestic violence, including intent, a burden the Agency failed to meet. Reply Br. 20-21; see Pet’r’s Br. 44, 49. Long contends that the Agency’s characterization of Long’s conduct in the specification of the Complaint establishes that it effectively charged Long with this criminal offense. Reply Br. 20-21. The Agency, however, explicitly charged Long with “conduct unbecoming an ALJ.” J.A. 72. “[W]hen an agency uses such general charging language, the Board must look to the specifica- tion to determine what conduct the agency is relying on as the basis for its proposed disciplinary action.” Russo v. U.S. Postal Serv., 284 F.3d 1304, 1308 (Fed. Cir. 2002). The specification of the Amended Complaint includes a narrative of the January 27, 2008 Incident, which refer- ences the criminal charges filed against Long, including domestic violence battery and culpable negligence. J.A. 72; J.A. 74-77. Yet the Amended Complaint expressly states that “the Agency seeks [Long]’s removal based upon his underlying conduct, not the fact that criminal charges have been brought against him. Regardless of the outcome of the criminal proceedings, a review of [Long]’s actions clearly show he is not fit to continue to serve as an ALJ.” J.A. 74; see J.A. 77. Therefore, despite referencing the criminal charges filed against Long, the Amended Complaint makes clear that the Agency’s charge of conduct unbecoming an ALJ is based on Long’s underlying conduct during the January 27, 2008 Incident, not the merits of the criminal charges. See Larry v. Dep’t of Justice, 76 M.S.P.R. 348, 355 (1997). The Board thus properly determined that the Agency was not required to LONG v. SSA 12 prove the elements of misdemeanor domestic violence. Final Decision at 21-22. As such, we conclude that substantial evidence sup- ports the Board’s finding that the Agency proved both specifications of the charge of conduct unbecoming an ALJ. II Pursuant to 5 U.S.C. § 7521, the Agency was permit- ted to remove or suspend Long, an ALJ, “only for good cause established and determined by the Merit Systems Protection Board.” 5 U.S.C. § 7521(a)-(b); Brennan v. Dep’t of Health & Human Servs., 787 F.2d 1559, 1561 (Fed. Cir. 1986). In addressing “good cause,” the Board noted that the ALJ position is one of “prominence,” meant to engender “great respect.” Final Decision at 20. The Board therefore held that an ALJ must not conduct himself in a manner that “undermines public confidence in the administrative adjudicatory process.” Id. The Board explained that it has held that the American Bar Association Model Code of Judicial Conduct (“ABA Model Code”) is an appropriate guide for evaluating the conduct of ALJs and quoted Canon 1, Rule 1.2 of the ABA Model Code: “A judge shall act at all times in a manner that promotes public confidence in the independence, integrity, and impartiality of the judiciary, and shall avoid impro- priety and the appearance of impropriety.” Id. at 20-21. Further, the Board explained that, in contrast to the “efficiency of the service” standard of 5 U.S.C. § 7513(a), “good cause” under 5 U.S.C. § 7521(a) does not require a “separate analysis of nexus.” Id. at 22. Turning to the facts of this case, the Board found that Long’s “physical altercation with his domestic partner, 13 LONG v. SSA resulting in the involvement of neighbors and the inter- vention of police officers . . . was inconsistent with main- taining respect for the administrative adjudicatory process.” Id. at 21. The Board found that this conduct meets the “good cause standard for disciplinary action.” Id. at 23; see id. at 21. A On appeal, Long does not contest the Board’s “good cause” standard and conceded, at oral argument, that the standard is acceptable and appropriate. Oral Arg. at 2:28-2:58, available at http://oralarguments.cafc.uscourts.gov/Audiomp3/2010- 3108.MP3. See generally Pet’r’s Br.; Reply Br. In contrast, Amicus Curiae, the Association of Administrative Law Judges (“AALJ”), objects to the Board’s construction of “good cause” and disputes the Agency’s argument that the Board’s interpretation is entitled to Chevron deference. Amicus Br. 9-24. We therefore address the Board’s inter- pretation of “good cause” in 5 U.S.C. § 7521. The Administrative Procedures Act (“APA”) does not define “good cause.” See Brennan, 787 F.2d at 1561. Indeed, we have recognized that “Congress intentionally failed to define ‘good cause’ in the . . . Act. Rather, ‘good cause’ is to be given meaning through judicial interpreta- tion . . . .” Id. at 1561-62. Although the “‘good cause’ standard was left to the courts to define,” courts have not provided a “succinct definition for ‘good cause.’” Id. at 1562. Courts have, however, articulated “what ‘good cause’ is not.” Id. at 1563. The Supreme Court, in Ramspeck v. Federal Trial Examiners Conference, 345 U.S. 128 (1953), held that “good cause” is not equivalent to the “good behavior” standard applicable to Article III judges. Brennan, 787 LONG v. SSA 14 F.2d at 1562. In addition, this court has held that a charge cannot constitute “good cause” if it is “based on reasons which constitute an improper interference with the ALJ’s performance of his quasi-judicial functions.” Id. at 1563. These holdings help narrow and frame the “good cause” standard but do not define “good cause.” In the absence of a controlling interpretation of “good cause” by the Supreme Court or this court, the Agency argues that we must give Chevron deference to the Board’s interpretation of “good cause.” Resp’t Br. 42-47. We agree. In Tunik v. Merit Systems Protection Board, 407 F.3d 1326 (Fed. Cir. 2005), we addressed the issue of whether the Board’s interpretation of 5 U.S.C. § 7521 is entitled to Chevron deference. We explained that the Supreme Court has “recognized a very good indicator of delegation meriting Chevron treatment i[s] express con- gressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.” Id. at 1336 (quoting United States v. Mead Corp., 533 U.S. 218, 229 (2001)). We concluded that 5 U.S.C. § 1305, which states that “for the purpose of section 7521 of this title, the . . . Board may investigate, prescribe regulations, appoint advisory committees as necessary, [and] recommend legislation,” provides such express congressional authori- zation for the Board to engage in rulemaking with respect to section 7521. Id. We further determined that “section 7521 authorizes the Board to adjudicate whether an agency has established good cause for disciplinary action against an ALJ.” Id. Thus, we held that “[t]he Board has been charged with administering section 7521 through both rulemaking and adjudication and is entitled to Chevron deference in these activities.” Id. Accordingly, under Tunik, the Board’s interpretation of “good cause” in section 7521 is subject to Chevron deference. 15 LONG v. SSA The AALJ, however, contends that the Board’s inter- pretation of “good cause” in section 7521 is not entitled to Chevron deference because courts have left open the question of whether Chevron deference is applicable where more than one agency is responsible for interpret- ing and implementing a statute. Amicus Br. 9 n.2. For support, the AALJ points to the Office of Personnel Man- agement’s (“OPM’s”) role in interpreting and rulemaking with respect to 5 U.S.C. § 7513(a), which includes the “efficiency of the service” standard. Id. at 9-10 n.2. It is true that courts have addressed but left unresolved the issue of whether Chevron deference is appropriate where multiple agencies are responsible for administering a statute. Bragdon v. Abbott, 524 U.S. 624, 642 (1998); Jones v. Dep’t of Transp., 295 F.3d 1298, 1307 n.5 (Fed. Cir. 2002). We conclude, however, that this open question does not affect the applicability of Chevron deference to the Board’s interpretation of 5 U.S.C. § 7521 because the Board has exclusive rulemaking and adjudicatory author- ity with respect to section 7521. See 5 U.S.C. §§ 1305, 7521; Tunik, 407 F.3d at 1336. OPM’s role in interpreting section 7513 is irrelevant. As such, our review of the Board’s interpretation of “good cause” in section 7521 is governed by Chevron. Under the two-part test of Chevron, a court must first determine “whether Congress has directly spoken to the precise question at issue.” Sullivan v. Everhart, 494 U.S. 83, 89-90 (1990). If so, that is the end of the inquiry because “the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Id. at 89. If, however, “the statute is silent or ambiguous with respect to the specific issue,” the court must sustain the agency’s construction if it is a “permissible construc- tion of the statute.” Id.; see Yellow Transp., Inc. v. Mich., 537 U.S. 36, 45 (2002). Specifically, the court must decide LONG v. SSA 16 “whether the agency’s construction is ‘rational and consis- tent with the statute.’” Sullivan, 494 U.S. at 89. “If the agency’s reading fills a gap or defines a term in a reason- able way in light of the Legislature’s design,” a court must give “that reading controlling weight, even if it is not the answer ‘the court would have reached if the question initially had arisen in a judicial proceeding.’” Regions Hosp. v. Shalala, 522 U.S. 448, 457 (1998). Here, it is clear that Congress has not “directly spo- ken to the precise question at issue,” namely the meaning of “good cause” in section 7521, as “Congress intentionally failed to define” the term. Brennan, 787 F.2d at 1561-62. Because section 7521, as well as the rest of the APA, is “silent or ambiguous” regarding the meaning of “good cause,” we must evaluate whether the Board’s construc- tion of “good cause” is “permissible.” We conclude that the Board’s interpretation of “good cause” to encompass conduct that “undermines public confidence in the admin- istrative adjudicatory process,” as informed by the ABA Model Code, is a permissible construction of the statutory language. See Final Decision at 20-22. In our view, the Board’s construction is rational, consistent with the APA, and reasonable in light of the APA’s design. We do not agree with the AALJ’s assertion that the Board’s inter- pretation of “good cause” prohibits ALJs from acting with the independence required by the APA. Amicus Br. 9 n.2, 13-14. The APA does indeed have provisions to ensure the “decisional independence” of ALJs and prohibits “substantive reviews and supervision of an ALJ’s . . . quasi-judicial functions.” Brennan, 787 F.2d at 1562. Yet the Board’s interpretation of “good cause” to cover conduct that “undermines public confidence in the administrative adjudicatory process” is not inconsistent with or in con- flict with such independence. Accordingly, under Chev- 17 LONG v. SSA ron, we must uphold the Board’s interpretation of “good cause” in section 7521. In upholding the Board’s interpretation of “good cause,” we note that the Board’s “good cause” standard does not require a “separate analysis of nexus.” Final Decision at 22. Again, Long concedes this point. Pet’r’s Br. 55; Oral Arg. at 1:18-25. The AALJ, however, takes issue with the Board’s failure to require an agency to establish a nexus linking the proven misconduct to the ALJ’s judicial duties, or to the agency’s reputation and ability to discharge its mission. Amicus Br. 17-19, 22-24. For support, the AALJ largely relies on cases involving the “efficiency of the service” standard of 5 U.S.C. § 7513(a). Section 7513(a) provides that an agency may take certain disciplinary actions against an employee “only for such cause as will promote the efficiency of the service.” 5 U.S.C. § 7513(a). We have held that this “efficiency of the service” standard requires that there be a “nexus between the employee’s misconduct and the work of the agency, i.e., the agency’s performance of its functions.” Doe v. Dep’t of Justice, 565 F.3d 1375, 1379 (Fed. Cir. 2009). The “efficiency of the service” standard of section 7513(a), however, is distinct from the “good cause” standard of section 7521(a), applicable to ALJs like Long. See 5 U.S.C. §§ 7513(a), 7521(a). The Board has repeatedly held, as it did here, that the “good cause” standard of section 7521 is not equivalent to the “effi- ciency of the service” standard of section 7513. Final Decision at 22-23; see, e.g., Soc. Sec. Admin. v. Carr, 78 M.S.P.R. 313, 338 (1998), aff’d, 185 F.3d 1318 (Fed. Cir. 1999); Soc. Sec. Admin. v. Mills, 73 M.S.P.R. 463, 467-68 (1996), aff’d, 124 F.3d 228 (Fed. Cir. 1997) (table). Thus, although a finding of nexus is required under the “effi- ciency of the service” standard of section 7513, the Board’s LONG v. SSA 18 failure to require a “separate analysis of nexus” for “good cause” under section 7521 was not error. Further, we reject the arguments put forward by the AALJ and to a lesser extent, Long, that the Board’s failure to require a “nexus” analysis for “good cause” under section 7521 allows ALJs to be disciplined for misconduct entirely unrelated to an ALJ’s judicial duties or the relevant agency. See Amicus Br. 8-12, 18; see also Pet’r’s Br. 54; Reply Br. 24, 27. Specifically, the AALJ contends that the Board’s “good cause” standard allows an agency to police an ALJ’s private, immoral behavior as well as his personal lifestyle choices, which have no impact on the agency or the ALJ’s ability to perform his duties. Amicus Br. 8-14, 18. These arguments overlook the substance of the Board’s “good cause” standard, which encompasses only conduct that “undermines public confi- dence in the administrative adjudicatory process.” Final Decision at 20. This standard ensures that misconduct constituting “good cause” for disciplinary action relates in some way to the character traits expected of an ALJ or to the agency’s mission generally. Indeed, misconduct wholly unrelated to an ALJ’s position, the characteristics expected of an ALJ, or the agency’s performance and mission would not in any way “undermine public confi- dence in the administrative adjudicatory process.” As such, despite the lack of a separate nexus requirement, the Board’s interpretation of “good cause” ensures that an ALJ’s misconduct must have some relationship to the agency or to the ALJ’s position in order to subject the ALJ to discipline. Thus, we uphold the Board’s interpretation of “good cause” in section 7521(a). 19 LONG v. SSA B Both Long and the AALJ object to the Board’s appli- cation of its “good cause” standard to the facts of this case, arguing that the Board erred in finding that Long’s con- duct was inconsistent with maintaining confidence in the administrative adjudicatory process. See Oral Arg. at 2:57-3:04, 16:40-19:35; Amicus Br. 11. We disagree. During the January 27, 2008 Incident, Long was physically violent with his domestic partner, Castro, both within and outside the home they shared. Long followed Castro, who held their child in her arms, into several different rooms of their house, escalating the incident with repeated physical assaults. Further, when Castro fled the house on foot, Long pursued her. As a result, the argument and Long’s physical violence against Castro spread into the neighborhood. The incident ultimately resulted in the involvement of neighbors and the inter- vention of police. Despite the AALJ’s attempts to classify Long’s mis- conduct as purely immoral, Long’s actions on the night of January 27, 2008 were violent, abusive, and potentially criminal. See Amicus Br. 8-9, 11-12, 18. Such violence is undoubtedly inconsistent with maintaining respect for and confidence in the administrative adjudicatory proc- ess. Long’s behavior calls into question his ability to perform his duties as an ALJ and to uphold the reputa- tion of the Agency, as it raises serious doubts regarding his possession of characteristics essential to the ALJ position, including judicial temperament, demeanor, control, and judgment. See Final Decision at 24. Thus, substantial evidence supports the Board’s finding that Long’s conduct undermined public confidence in the administrative adjudicatory process. LONG v. SSA 20 In so holding, we reject Long’s and the AALJ’s argu- ment that the January 27, 2008 Incident did not result in sufficient notoriety or public awareness to warrant “good cause” for disciplinary action against Long. Pet’r’s Br. 54; Amicus Br. 8, 12; Oral Arg. at 16:40-19:35. As the Board recognized, the January 27, 2008 Incident did not receive press coverage or substantial public notoriety. Final Decision at 27 & n.13. Nevertheless, the record belies Long’s and the AALJ’s suggestion that the incident was a private matter about which the public was unaware. The January 27, 2008 Incident spread into the neighborhood when Castro left the house and Long followed her in his car. The verbal argument and Long’s physical violence against Castro continued in public view. At least two neighbors, the Feeneys, who lived almost a mile from Long and Castro’s home, became actively involved in the incident when Castro arrived at their home, frightened and seeking help. Moreover, after the police intervened and arrested Long, the police posted his mug shot on the publicly accessible Broward County Sheriff’s Office’s website. J.A. 306-07; J.A. 1359; J.A .1475; Oral Arg. at 18:36-45. Long’s co-workers and other ALJs were also aware of the incident. J.A. 1110. Further, we disagree with the AALJ’s implication at oral argument that an ALJ’s misconduct must actually receive negative publicity, e.g., through the newspaper, television, or the internet, in order to raise concerns regarding public confidence in the administrative adjudi- catory process. See Oral Arg. at 16:40-19:35. Such press coverage is not required for an ALJ’s conduct to under- mine public confidence in the administrative adjudicatory process. The propriety of disciplinary action against an ALJ does not rest on the happenstance of media attention regarding the ALJ’s behavior. 21 LONG v. SSA Additionally, both Long and the AALJ dispute the Board’s finding of “good cause” on the grounds that there was no evidence that Long’s conduct actually impacted his workplace or ability to perform his duties. Pet’r’s Br. 54; Reply Br. 24; Amicus Br. 11-12, 16. The Board’s “good cause” standard, however, requires only that the ALJ’s conduct undermine confidence in the administrative adjudicatory process. Such an undermining of confidence occurs where the conduct creates doubts in the ALJ’s ability to carry out his responsibilities or raises concerns that the ALJ’s behavior will reflect poorly on the agency and its adjudicatory process. Here, Long’s supervisors, including the Chief ALJ and the Regional Chief ALJ, testified that his behavior caused them to have serious concerns regarding his ability to adequately perform his duties. Final Decision at 24. Moreover, the fact that the incident was not purely private raised realistic concerns that the public would share such doubts as to Long’s ability and the propriety of the Agency’s adjudicatory process. See id. at 21 n.9 (quoting Lisa Feeney and Don- ald Feeney’s impressions regarding Long’s conduct in relation to his position as an ALJ). As such, we uphold the Board’s finding that Long’s conduct during the Janu- ary 27, 2008 Incident constituted “good cause” for disci- plinary action against him. III Finally, Long argues that the court should mitigate the penalty of removal because the Board misapplied the Douglas factors and removal is so unconscionably dispro- portionate to Long’s conduct that it amounts to an abuse of discretion. Pet’r’s Br. 57-72; Reply Br. 29-35. We may overturn a penalty imposed by the Board for an ALJ’s misconduct “[o]nly in the exceptional case in which the penalty exceeds that permitted by statute or regulations LONG v. SSA 22 or is so harsh that it amounts to an abuse of discretion.” Brennan, 787 F.2d at 1563. Here, the Board conducted a thorough analysis of the Douglas factors and found the nature and seriousness of Long’s misconduct, as well as its negative impact on the Agency and his supervisors’ confidence in his ability to adequately perform his judicial duties, warranted removal. Final Decision at 24-27. Particularly in light of the gravity of the sustained mis- conduct and evidence that this conduct caused Long’s supervisors to have serious concerns regarding his ability to effectively serve as an ALJ, we cannot conclude that the imposed penalty of removal was so harsh that it amounts to an abuse of discretion. Accordingly, we up- hold the Board’s penalty of removal. AFFIRMED United States Court of Appeals for the Federal Circuit __________________________ DANVERS E. LONG Petitioner, v. SOCIAL SECURITY ADMINISTRATION Respondent. __________________________ 2010-3108 __________________________ Petition for review of the Merit Systems Protection Board in case no. CB7521080019-I-1. __________________________ Dyk, Circuit Judge, concurring. While I join the majority opinion, I write separately to express my concerns about the claimed role of the Merit Systems Protection Board (“Board”) in determining whether federal employees have engaged in misconduct outside the workplace. 1 Whether the standard is the usual “efficiency of the service” standard, 5 U.S.C. § 7513, or the “good cause” standard governing discipline for administrative law judges (“ALJs”), 5 U.S.C. § 7521, I doubt that Congress intended to broadly authorize the 1 In addition to this case, our opinion in Doe v. De- partment of Justice, 565 F.3d 1375 (Fed. Cir. 2009), involving alleged off-duty misconduct by a federal law enforcement officer, raised similar issues. LONG v. SSA 2 Board to conduct investigations into federal employees’ private conduct. Neither the Board nor federal agencies are given a general warrant to ferret out misconduct in the private lives of federal employees. It is one thing for the Board to sustain the dismissal of an employee based on a criminal conviction; it is quite another for the Board to adjudicate criminal behavior when state or federal authorities have declined prosecution. Despite the Board’s disclaimers, that is effectively what the Board did here; it adjudicated criminal conduct. The procedures and evidentiary standards were, however, substantially different than in criminal proceedings. The petitioner was not afforded a jury trial. Moreover, most of the evidence on which the Board relied would have been inadmissible in a criminal trial. Not only does most of the evidence (e.g., the transcript of Ms. Castro’s statement on the night of the incident and the testimony by the two police officers and Mr. and Mrs. Feeney about what Ms. Castro told them happened on the night of the incident) constitute hearsay, but the critical item of hearsay––the transcript of the police interview with Ms. Castro on the night of the incident––was not even authenticated, see Fed. R. Evid. 901. On top of this, the Board rejected some of the factual findings of the administrative judge who, unlike the Board, heard live witness testimony. Although these failings do not constitute reversible error in this case, they certainly raise concerns about fairness in these types of proceedings. Lastly, neither the Office of Personnel Management nor the Board has articulated any consistent or compre- hensive standard for when private, off-duty actions may lead to workplace discipline, whether they constitute “good cause” or satisfy the efficiency of the service stan- dard. See Doe, 565 F.3d at 1380–81. In Doe, we over- turned the Board’s decision sustaining the removal of a 3 LONG v. SSA federal law enforcement officer for “clearly dishonest” conduct because the Board “failed to articulate a mean- ingful standard as to when private dishonesty rises to [a] level of misconduct” where discipline is warranted under the “efficiency of the service” standard. Id. at 1380. We explained that “without a predetermined standard [con- cerning when off-duty misconduct can subject employees to discipline] . . . federal employees are not on notice as to what off-duty behavior is subject to investigation and the government could use this overly broad standard to legitimize removals made for personal or political rea- sons.” Id. at 1381. Although the standards issue was not properly raised by the petitioner in this case, these same concerns also apply to the removal of ALJs for off-duty misconduct. These features of the present case––the criminal na- ture of the charges, the prevalence of hearsay and non- authenticated evidence, the rejection of the administra- tive judge’s findings, and the lack of a comprehensive standard––are disquieting. There is room for such Board action in unusual cases, and this appears to be one of those cases. This is so because of the seriousness of the charge, the fact that the altercation spurred police in- volvement, the strength of the hearsay evidence, the relatively small number of differences between the Board and administrative judge on their views of the facts, and the failure of Long to raise the standards issue. However, the Board must engage in such proceedings only in the most unusual circumstances or risk reversal by this court.