CUNA/LEAGUE 2015 EXAM SURVEY REPORT February 2016

CUNA/LEAGUE 2015 EXAM SURVEY REPORT February 2016 Prepared by: Mike Schenk Connie Dey-Marcos Jon Haller

SURVEY METHODS and RESPONDENT PROFILE In October, 2015, CUNA s News Now included an article reminding credit unions to complete the CUNA Exam Survey. This was followed by several e-mails sent directly to individual credit unions urging them to complete the survey for the most recent exam. The questionnaire was almost identical to one used a year earlier. By January, 2016, we received 657 responses, representing 10% of all credit unions. The table below compares the respondent profile to that of all U.S. credit unions. The distribution of responding credit unions is very similar to that of the population in terms of charter type. However, responding credit unions were somewhat larger than all US credit unions: 22% of responding credit unions have more than $250 million in assets compared to 14% of the population. Nevertheless, there was strong response across all asset sizes. Because larger credit unions were more likely to respond, responses from single common bond credit unions were lower than the population, and community charters were more heavily represented. All totals reported in the survey have been weighted to the distribution of all credit unions by asset size. RESPONDENT PROFILE Survey Respondents All Credit Unions Number of Credit Unions 650 6,329 Charter State 39% 39% Federal 61% 61% Field of Membership Single common bond 23% 34% Multiple common bond 32% 32% Community 45% 34% Asset Group $25 million or less 38% 49% $25 million to $50 million 14% 14% $50 million to $250 million 27% 24% $250 million to $500 million 9% 5% $500 million to $1 billion 4% $1 billion or more 7% 4% Net Worth Ratio Less than 6.00% 1% 1% 6.00% to 6.99% 2% 2% 7.00% to 9.99% 3 30% 10.00% or greater 61% 68% Page 2

MAJOR FINDINGS Once again, more credit union CEOs are satisfied with their exams (65%) than dissatisfied (21%). This finding is substantially improved compared to that reported last year (58%/28%). These are the highest levels of overall satisfaction seen in the four years this study has been conducted. Previously, the highest satisfaction was seen in 2012 when 61% were satisfied and 25% were dissatisfied. Exams duration is little-changed in 2015 (9.2 days). This is nearly identical to both the 2014 result (9.0 days) and the 2013 result (9.1 days) but a bit longer than that seen in 2012 (7.9 days). There was a significant decline in the proportion of credit unions reporting being under one or more Documents of Resolution (DORs). In 2015 30% of responding credit unions indicate they are under at least one DOR. In 2014 fully 40% of responding credit unions were under at least one DOR roughly the same as in 2013 (41%) but a bit lower than the finding in 2012 (43%). Exams conducted by state examiners remain substantially less likely to include DORs than exams in which NCUA is involved. Satisfaction with examinations conducted solely by state examiners tend to be rated similarly to those that are NCUA-only exams, but joint exams still are rated lower than either state or NCUA only exams. Once again, credit unions whose CAMEL ratings declined during the exam are not very positive about the experience. Similarly, as would be expected, there remains a strong, positive correlation between level of agreement with the current CAMEL rating and satisfaction with exam results. Exam teams again receive very positive ratings on giving credit unions the opportunity to comment, knowledge of rules and regulations, being open to discussion, and knowledge of the credit union. Exam teams continue to receive negative ratings on such things as applying "guidance" or "best practice" as regulation, and "covering" themselves. The biggest problem mentioned by credit unions is that regulatory and exam requirements in general are putting increasing pressure on credit union resources. ORGANIZATION OF THIS REPORT The report contains three pages each on a number of questions covering credit unions evaluations of the exam, the results of the exam, and assessments of the exam team in general, and in its dealings with volunteers. In each case, the first page contains a chart showing a comparison of responses to the question annually over the 2012-2015 period, for all respondents. The second page has a number of smaller charts showing how responses vary according to field of membership, examining agency, change in CAMEL rating, agreement with CAMEL rating, asset size, and net worth ratio. The last two tables list, in rank order, more detailed ratings of exam teams and the exam process, and issues or problems raised during exams. It is important to note that percentages may not add to 100% due to rounding. Also: Differences in the percent of credit unions responding to any question of 3 points or less are not statistically significant. For most questions, averages are calculated on a five-point scale. Once again, relatively few credit unions with low net worth ratios responded to the survey in 2015 only ten with less than net worth and only eleven with net worth between and 7% did so. As a consequence, the average results reported in each of these categories are idiosyncratic and are difficult to compare to prior year s results. Page 3

Overall Satisfaction with Exam 2012 Average = 2013 Average = 2014 Average = 2015 Average = Overall Satisfaction With Exam 7% 10% 10% 10% 15% 17% 18% 14% 15% 15% 14% 14% 21% 20% 24% 22% 40% 38% 34% 43% 1: Very dissatisfied 2: Somewhat Dissatisfied 3: Neutral 4: Somewhat satisfied 5: Very Satisfied 2012 (Avg=) 2013 (Avg=) 2014 (Avg=) 2015 (Avg=) More credit union CEOs are satisfied with their exams (65%) than dissatisfied (21%). These are the highest levels of overall satisfaction seen in the four years this study has been conducted. Previously, the highest satisfaction was seen in 2012 when 61% were satisfied and 25% were dissatisfied. Credit unions with a single common bonds or with multiple common bonds are more satisfied than those with multiple common bonds, though all three groups refelct higher satisfaction in 2015. In 2014 credit unions were more satisfied with exams performed by state regulators than by NCUA. However, the 2015 survey shows no difference in the average level of satisfaction between state and NCUA exams. In contrast, joint exams reflect substantially lower levels of satisfaction, and satisfaction with joint exams is essentialy unchanched after improving slightly in both 2013 and 2014. Not surprisingly, satisfaction is substantially higher for exams in which CAMEL ratings improved, and vice versa. Satisfaction is also once again strongly related to level of agreement with the CAMEL rating. There is little variation in average satisfaction by asset size, however, those with $500 million to $1 billion in assets reflect the highest level of satisfaction and those in the $50 million to $250 million category reflect the lowest level of satisfaction. Average satisfaction among those with $1 billion or more in assets declined slightly in 2015, while those with less than $25 million reflect no change during the year. In contrast, each of the other four asset categories reflects increases during the period. Credit unions with 7% to 10% net worth as well as those with 10% or higher net worth reflect modestly higher levels of satisfaction in 2015. Page 4

Overall Satisfaction with Exam 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.5 2.3 2.2 2.3 1.5 1.5 1.5 1.5 2.5 2.5 2.5 2.4 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 1.33 2.9 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 5

Change in Camel Rating 1 2012 Average = 2013 Average = 2014 Average = 2015 Average = Change in CAMEL Rating 11% 12% 72% 72% 75% 74% 12% 11% 17% 15% 14% 15% 1: Declined 3: Remained the Same 5: Improved As in prior years, the vast majority of credit unions have unchanged CAMEL ratings in their latest exams. There is a modest increase in the proportion reporting a rating improvement in 2015, and a marginal decline in the proportion reporting both a lower rating and a rating that remained the same. Single and community common common bond credit unions are more likely to experience increases in CAMEL ratings in the current cycle though the differences between common bond groups is relatively small. There is very little difference in average reported change in CAMEL rating according to the examining agency, though those with dual exams report a modestly higher improvement than did their counterparts that were examined only by state regulators or only by NCUA. As would be expected, the greater the level of agreement with the CAMEL rating, the more likely is the CAMEL rating to have increased a result that is similar to that reported in previous survey result summaries. In 2015, credit unions with $500 million or more in assets were most likely to see their CAMEL ratings increase. None in the $500 million to $1 billion asset group reported that their CAMEL rating declined. Credit unions with 7% to 10% capital ratios were most likely to see improved CAMEL ratings during 2015, though those with less than 7% capital experienced the biggest positive changes in the year. 1 Averages calculated on a scale: 1 = declined; 3 = remained the same; 5 = improved. Page 6

Change in Camel Rating 2 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Agreement with CAMEL Rating 2.3 2.0 2.0 1.9 2.7 2.4 2.6 2.5 2.9 2.9 2.9 Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.3 2.3 2.8 2.9 2.7 2.8 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > 2 Averages calculated on a scale: 1 = declined; 3 = remained the same; 5 = improved. Page 7

Level of Agreement with Camel Rating 2012 Average = 2013 Average = 2014 Average = 2015 Average = Level of Agreement with CAMEL Rating 8% 7% 7% 5% 14% 12% 1 14% 17% 13% 14% 14% 23% 2 2 23% 39% 37% 41% 44% 1: Strongly 2: somewhat 3: Neutral 4: Agree Somewhat 5: Agree Strongly The overall average level of agreement with the CAMEL rating increased in 2015, with a fairly large increase in those saying they agree strongly with their rating. There is little difference in level of agreement with CAMEL ratings between credit unions with different fields of membership. However, community chartered credit unions reflect the largest increase in level of agreement during 2015 a tree-tenths of a point jump from in 2014 to in 2015. Agreement with CAMEL rating on exams by a state regulators are unchanged in 2015 but they continue to reflect the highest levels of agreement. Those with NCUA exams reported the most substantial increase in level of agreement, increasing from an average of in 2014 to in 2015. The level of agreement in NCUA exams was only in 2012. Those with joint exams once again reflect the lowest level of agreememnt with their CAMEL rating, though the average increased a bit in 2015. Once again credit unions whose CAMEL ratings improved are much more likely to agree with those CAMEL ratings. Larger credit unions were again slightly more likely to agree with their CAMEL ratings than smaller credit unions. Credit unions with very high capital ratios are more likely to agree with their CAMEL ratings than credit unions with lower capital ratios Page 8

Level of Agreement with CAMEL Rating 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Asset Size 2.3 2.4 2.3 2.3 4.5 4.4 Improved Remained the Same Declined By Net Worth Ratio 2.0 2.8 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 9

EXAM TEAM RATINGS As in the previous surveys, respondents were asked to rate their exam teams on a number of key attributes, including objectivity, fairness, helpfulness and professionalism. Exam teams again receive the highest ratings for professionalism and the lowest rating, on average, for objectivity. The average ratings for each of the four attributes studied increased modestly in the most recent survey. Average Examiner/Exam Team Ratings Professionalism Helpfulness Fairness Objectivity Page 10

Rating the Exam Team: Professionalism 2012 Average = 2013 Average = 2014 Average = 2015 Average = Professionalism 3% 4% 5% 2% 5% 8% 11% 13% 11% 12% 38% 39% 39% 42% 39% 45% 40% 31% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The average rating of examiner professionalism is little changed in 2015, though there was a five percentage point decline in the proportion of respondents rating examiner professionalism excellent. There also were declines in the proportion that rated examiner professionalism as poor or somewhat poor which decreased from 13% in 2014 to 8% in 2015. Single common bond credit unions reflect the highest average evaluations of examiner professionalism, though the differences by field of membership are not great. This result is consistent with findings in each of the previous three Exam Surveys. Reported examiner professionalism is lower on joint exams than either state only, or NCUA-only exams, and the average differnces between joint and other exams widened in the 2015 survey. Examiners who assigned reduced CAMEL ratings are viewed as significantly less professional than their peers who assigned higher ratings, at least in the eyes of the examined. This result has been consistent over time, though the average view of examiner professionalism among those in this group increased significantly in the year. In a similar vein, the greater the level of agreement with the CAMEL rating, the more likely the credit unions is to rate the examiners as highly professional. Again, as might be expected, this result has been consistently seen across in each of the four Exam Surveys we ve conducted. Perceptions of examiner professionalism are nearly identical across the six credit union asset-size categories examined. Credit unions with net worth ratios between and 7% are most likely to give high ratings for professionalism and the average rarings in this gropu have increased in each of the past three years. Page 11

Rating the Exam Team: Professionalism 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 4.4 2.6 2.8 2.8 4.6 4.4 4.5 4.5 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 4.4 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 12

Rating the Exam Team: Helpfulness 2012 Average = 2013 Average = 2014 Average = 2015 Average = Helpfulness 5% 5% 7% 4% 7% 8% 15% 17% 17% 1 35% 38% 30% 38% 39% 33% 38% 35% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent Views of examiner helpfulness improved marginally in 2015, reflected in an average rating of on a five point scale up from an average of in 2014. Results reflect a five percentage point increase in the proportion of respondents rating examiner professionalism good or excellent. At the other end of the spectrum there are declines in the proportion that rated examiner professionalism as poor, or somewhat poor as well as a slight decline among those who were neutral on the subjet. In each of the four years we ve conducted the Exam Survey single common bond credit unions had higher evaluations of examiner helpfulness that did multiple bond or community credit unions. Reported examiner helpfulness is once again noticeably lower on joint exams than either state-only, or NCUA-only exams. State exams continue to reflect the highest perceptions of examiner helpfulness. Examiners who assigned reduced CAMEL ratings are deemed to be less helpful than their peers who assigned higher ratings. This result is consistent with previous findings. Similarly, once again, the greater the level of agreement with the CAMEL rating, the more likely the credit union is to rate the examiners as helpful. The nation s smallest credit unions those with less than $25 million in assets - were again the most likely to rate the exam teams as helpful. Perceptions of examiner helpfulness generally decline as credit union asset size increases. Ratings for examiner helpfulness are relatively low among credit unions with net worth ratios below. Page 13

Rating the Exam Team: Helpfulness 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.9 2.9 2.2 2.5 2.4 4.5 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.0 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 14

Rating the Exam Team: Fairness 2012 Average = 2013 Average = 2014 Average = 2015 Average = Fairness 5% 5% 8% 8% 11% 9% 1 17% 14% 14% 3 39% 34% 43% 34% 31% 34% 29% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The average rating of examiner fairness is unchanged in 2015 at on the five-point scale. The proportion of respondents rating examiner fairness as excellent in 2015 is five percentage points lower than that found in 2014. However, those rating fairness as good increased by nine percentage points in the current survey cycle. In addition, the proportion of those rating examiner objectivity as poor or somewhat poor in 2015 is a combined three percentage points lower than that seen in 2014. Single common bond credit unions are more likely to describe examiners as fair than are other credit unions. This result is consistent with findings in previous years. Once again, examiners in state-only exams are slightly more likely to be rated as fair than in NCUA-only exams while examiners in joint exams are much less likely to be considered fair than in either of the solo exams. The better the change in CAMEL rating, the more likely is the exam team to be considered fair and the differences seen in 2015 are a bit less pronunced than those observed in previous survey cycles. The more credit unions agree with their CAMEL ratings, the more likely they are to consider the examiner fair. These results are consistent over time. Credit unions with less than $50 million in assets and those with $1 billion or more in assets are most likely to describe the exam team as fair. Credit union perception of examiner fairness increases with reported capital ratio. Page 15

Rating the Exam Team: Fairness 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 1.9 1.9 1.9 2.7 2.8 2.5 2.8 4.5 4.4 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.0 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 16

Rating the Exam Team: Objectivity 2012 Average = 2013 Average = 2014 Average = 2015 Average = Objectivity 7% 9% 10% 10% 12% 10% 1 18% 15% 17% 3 37% 31% 39% 32% 28% 33% 28% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The average rating of examiner objectivity is unchanged in 2015 at on the five-point scale. The proportion of respondents rating examiner objectivity as excellent in 2015 is five percentage points lower than that found in 2014. However, those rating fairness as good increased by eight percentage points in the current survey cycle. Additionally, the proportion of those rating examiner objectivity as poor or somewhat poor in 2015 is a combined five percentage points lower than that seen in 2014. Ratings for examiner objectivity are highest among single common bond credit unions, and lowest among community credit unions. Credit unions with state-only exams rate objectivity highest, followed by those with NCUA-only exams, but those with joint exams rate examiner objectivity much lower than their counterprts. Examiners who assigned a lower CAMEL rating than the previous exam are considered to be much less objective a finding that has been consistent over time. Once again, the more a credit union agrees with its CAMEL rating, the more likely it is to rate the exam team as objective. Credit unions with assets below $25 million rate examiners as more objective than other credit unions. Perceptions of examiner objectivity are lowest among credit unions with $50 to $250 million in assets. By net worth ratio, the highest ratings for objectivity are seen in credit unions with ratios above 10%. Page 17

Rating the Exam Team: Objectivity 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 1.9 1.7 1.7 2.9 2.7 2.7 2.4 2.8 2.7 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.3 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 18

EXAM TEAM DEALINGS WITH VOLUNTEERS RATINGS As in each of the previous surveys, respondents were asked to rate their exam teams on their dealings with the board of directors/volunteers. Key attributes evaluated included: willingness to account for the credit union s business plan when discussing regulatory exceptions; willingness to share agency insight if the report called for corrective action(s) to be taken; dissemination of the report to the board/supervisory committee; and openness to interactive dialogue regarding the report. Average ratings for exam team dealings with volunteers improved marginally (though not dramatically) in the most recent survey. Exam team average ratings on willingness to share insights on corrective action declined but only slightly in 2015, while average ratings on each of the other three key attributes were a bit higher compared to 2014 results. In 2015, exam teams again received the lowest ratings for willingness to account for the credit union s business plan when discussing regulatory exceptions. Dealings with Volunteers Ratings Openness to dialogue on report Dissemination of the report Willingness to share insight on corrective actions Willingness to account for CU business plan Page 19

Rating the Exam Team s Dealing With Volunteers: Dissemination of Report 2012 Average = 2013 Average = 2014 Average = 2015 Average = Dissemination of Report 3% 4% 4% 3% 7% 17% 19% 18% 18% 43% 41% 39% 40% 31% 30% 32% 33% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The average rating of the exam team s dissemination of the exam report is unchanged in 2015. CEOs at community-based credit unions are a bit less likely than other CEOs to give high ratings for dissemination of the exam report to volunteers. Ratings on dissemination of the exam report are lowest at credit unions with joint exams and the highest ratings by examining agency are for state-only exams. This result is consistent with findings in each of the past three survey cycles. Credit unions whose CAMEL rating declined are more likely to give very low ratings for dissemination of the report. Once again, the more a credit union agrees with its CAMEL rating, the more highly it tends to rate the exam team for dissemination of the exam report. There is very little variance in credit union rating of exam team s dissemination of exam report by asset size, though credit unions with less than $25 million in assets once again reflect the highest ratings on this metric. Credit unions with net worth ratios of 10% or higher view report dissemination more highly than their counterparts with lower ratios. Page 20

Dealing with Volunteers- Dissemination of the Report 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.6 2.8 2.6 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 4 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 21

Rating the Exam Team s Dealing With Volunteers: Openness to Dialogue 2012 Average = 2013 Average = 2014 Average = 2015 Average = Openness to Dialogue 3% 4% 4% 4% 7% 17% 19% 18% 15% 43% 41% 39% 38% 31% 30% 32% 37% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The average rating of the exam team s openness to dialogue regarding the report is little-changed in 2015. However, a substantial increase in the percentage of credit unions calling the openness to dialouge excellent is evident. By charter-type, single common bond credit unions give the highest ratings for openness to dialog, though ratings among those with multiple group and commumity charters were not substantially lower than those seen in the single bond group. State-only and NCUA-only exams result in higher ratings for openness to dialog with volunteers than did joint exams. Credit unions with CAMEL ratings that declined in the cycle are substantially less likely to credit examiners with being open to dialog than are other credit unions a result reported in previous survey cycles as well. Once again, the more a credit union agrees with its CAMEL rating, the more likely it is to rate the exam team as open to dialog with volunteers. Credit unions with assets of $25 million or less gave the highest ratings for openness to dialog. Credit union perceptions of openness to dialog tends to increase with net worth ratio. Page 22

Dealing with Volunteers- Openness to Dialogue 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.5 2.6 2.5 4.5 4.4 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 23

Rating the Exam Team s Dealing With Volunteers: Willingness to Share Insight 2012 Average = 2013 Average = 2014 Average = 2105 Average = Willingness to Share Insight 3% 4% 4% 3% 5% 7% 20% 22% 18% 21% 38% 39% 39% 37% 33% 30% 32% 32% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent Ratings of the exam team s willingness to share agency insight in dealing with directors/volunteers has averaged on a 5-point scale in each of the past four years. Single common bond credit unions rate examiners a bit more highly than multiple-group or community credit unions on examiner willingness to share insight. Examiners on state-only exams and in NCUA-only exams received the highest ratings for willingness to share insight, while examiners on joint exams received the lowest ratings. NCUA-only exams refelct an increase in their average ratings over compared to 2014 results. As might be expected, credit unions with CAMEL ratings that declined in the cycle once again give substantially lower marks for willingness to share insight than do those with ratings that did not decline. The more a credit union agrees with its CAMEL rating, the more likely it is to rate examiners highly for willingness to share insight. Ratings for willingness to share insight are highest among credit unons with less than $25 million in assets. Average ratings for willingness to share insight tends to increase with net worth ratio category. Page 24

Dealing with Volunteers- Willingness to Share Insight 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.6 2.6 2.5 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.3 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 25

Rating the Exam Team s Dealing With Volunteers: Willingness to Account for CU Business Plan 2012 Average = 2013 Average = 2014 Average = 2015 Average = Willingness to Account for CU Business Plan 5% 5% 5% 7% 7% 7% 24% 25% 25% 25% 35% 35% 33% 33% 30% 27% 29% 30% 1: Poor 2: Somewhat Poor 3: Neutral 4: Good 5: Excellent The rating of the exam team s willingness to account for the credit union s business plan in dealing with directors/volunteers increased marginally - from an average in 2014 to an average of in 2015. Credit unions with single and multiple common bonds give higher marks for willingness to account for the credit union's business plan than did those with community charters. Ratings for willingness to consider the credit union's business plan are highest for NCUA-only exams and both NCUA-only and state-only exams are rated higher than joint exams. Examiners who assign reduced CAMEL ratings are once again judged to be substantially less willing to account for the credit union's business plan. The more a credit union agrees with its CAMEL rating, the higher the rating it gives for examiner willingness to account for the credit union's business plan. Credit unions with assets below $25 million in assets are most likely to give high marks for examiner willingness to account for the credit union's business plan. Views of examiner willingness to account for a credit union's business plan tends to increase with net worth ratio. Page 26

Dealing with Volunteers- Willingness to Account for Business Plan 2012 Average = 2013 Average = 2014 Average = 2015 Average = By Field of Membership By Examining Agency Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 2.8 2.8 2.2 2.3 2.3 2.9 4.4 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 2.0 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 27

Length of Exam: First Appearance to Completion 2012 Average = 7.9 days 2013 Average = 9.1 days 2014 Average = 9.0 days 2015 Average = 9.0 days First Appearance to Completion 13% 9% 8% 8% 12% 20% 20% 20% 20% 37% 32% 34% 31% 18% 18% 19% 21% 21% 19% 19% 1-3 Days 4-6 Days 7-9 Days 10-12 Days 13 or more Days Exam duration is unchanged in 2015 at 9.0 days though average duration remains elevated compared to the result we observed in 2012 (7.9 days). Community chartered credit union exams take longer than exams of multiple bond credit unions, which in turn take longer than exams of single common bond credit unions. This result is consistent with findings in each of the past three years. It should be noted that community chartered credit unions tend to be larger, on average, than their counterparts with other charters. Once again findings reflect the fact that joint exams took considerably longer than NCUA-only exams which took longer than state-only exams. Exams at which the CAMEL rating improved took longer than those for which the CAMEL rating remianed the same or improved. Exams about which the credit union disagreed most with the CAMEL rating took longer than other exams. The larger the credit union, the longer the exam. (Note: since our assumption for the number of days for exams lasting more than 12 days is the same for all such exams, our estimate of the difference in the length of exams for the two largest asset categories likely understates actual differences.) Exams of well-capitalized credit unions took longer than exams of those that were adequately or less than adequately capitalized. Page 28

Length of Exam: First Appearance to Completion 2012 Average = 7.9 days 2013 Average = 9.1 days 2014 Average = 9.0 days 2015 Average = 9.2 days By Field of Membership By Examining Agency 6.3 7.5 7.0 7.3 9.1 9.4 8.1 8.7 8.5 9.9 10.110.0 6.6 7.8 7.5 8.1 7.8 8.8 8.4 8.7 11.8 11.8 10.9 9.7 Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating 9.0 9.6 7.6 8.0 9.1 10.1 9.1 9.1 9.1 9.4 8.5 8.1 9.4 11.1 10.0 10.1 By Agreement with CAMEL Rating 8.5 9.5 9.3 10.1 6.6 8.3 8.4 7.6 7.9 9.5 8.8 9.2 8.1 8.9 8.9 9.3 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 5.8 6.2 5.9 6.2 8.3 9.6 8.3 8.6 8.3 9.6 10.0 10.1 12.0 1 1 12.8 12.5 1 1 1 12.1 14.5 15.0 14.9 9.0 7.6 8.0 6.9 6.0 6.3 5.7 5.0 9.6 8.7 9.19.0 9.29.19.3 7.5 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 29

Length of Exam: Completion to Delivery of Final Report 2012 Average = 1 days 2013 Average = 1 days 2014 Average = 1 days 2015 Average = 1 days Completion to Delivery of Final Report 7% 5% 7% 10% 9% 5% 11% 10% 11% 10% 12% 17% 15% 14% 12% 55% 60% 6 59% 1-3 Days 4-6 Days 7-9 Days 10-12 Days 13 or more Days The average length of time from completion to delivery of the final report declined, but only marginally in 2015. On average it took 1 days a bit lower than the 1 day duration reported in both 2013 and 2014. The time to delivery of the final report is highest at community credit unions, and lowest at single common bond credit unions a result consistently reported in each of the four survey cycles. The time to delivery of the final report is once again shortest at NCUA-only exams, and longest in joint exams. The time to delivery of the final report is longer at credit unions experiencing a decline in their CAMEL rating. In general, the more a credit union disagreed with its CAMEL rating, the longer is the time to delivery of the final report. This result is consistent across each of the four survey cycles. Larger credit unions had to wait longer for the delivery of the final report than smaller credit unions. Higher net worth ratios are generally refelctive of shorter times to delivery of the final report. Page 30

Length of Exam: Completion to Delivery of Final Report 2012 Average = 1 days 2013 Average = 1 days 2014 Average = 1 days 2015 Average = 1 days By Field of Membership By Examining Agency 12.7 1 1 12.6 1 1 14.5 1 1 1 15.3 1 15.6 14.9 14.9 14.9 11.8 12.5 1 12.4 15.3 16.4 16.0 16.5 Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 1 1 1 1 1 1 14.4 1 1 15.5 15.8 1 1 15.5 16.5 16.2 14.4 15.3 14.4 1 1 1 1 12.7 12.9 1 15.2 1 1 1 1 1 Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 12.1 12.6 1 12.3 1 1 1 12.8 1 1 14.7 1 15.0 15.4 16.3 14.9 14.5 15.9 16.2 15.3 14.5 16.4 16.2 17.1 10.0 15.2 18.0 14.7 11.2 14.8 14.5 1 1 1 14.9 1 1 1 1 1 <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 31

Written Agreements 3 2012 = 44% 2013 = 44% 2014 = 41% 2015 = 32% Written Agreements 4% 4% 4% 2% 1% 1% 1% 1% 2% 2% 1% 43% 41% 40% 30% 5 5 59% 68% DOR LUA NWRP Revised BP None The proportion of credit unions under any form of written agreement, while still relatively high, declined dramatically in 2015. Overall, 32% of recently-examined credit unions say they now face one or more written agreements down from 41% in the 2014 survey cycle. o The number of credit unions reporting Documents of Resolution (DORs) fell dramatically (from 40% in 2014 to 30% in 2015) o The prevalance of Revised Business Plans (BPs) also declined but only marginally (from 2% in 2014 to 1% in 2015). o Letters of understanding and agreement (LUAs) increased margianlly (from 4% in 2014 to in 2015) while the proportion under net worth restoration plans (NWRPs) is unchanged at 1%. The percentage of credit unions reporting some form of written agreement is similar across FOM-types, though community charters are more than two times as likely as ther couterparts to say they are under a LUA. Credit unions with dual exams are much more likely to report being under one or more written agreements than are credit unions with NCUA-only or state-only exams. Credit unions whose CAMEL rating worsened are very likely to report being under one or more DORs. In a strong relationship, the less satisfied a credit union is with its CAMEL rating, the more likely it is to be under one or more written agreements. Credit unions with less than $1 billion in assets are substantially more likely to be under written agreements than are larger credit unions. 3 Credit unions could report more than one type of written report. Note also that we ask only if one or more DORs was in effect, NOT the total number of DORs. Page 32

Written Agreements % of CUs Under One or More Written Agreements 2012 = 44% 2013 = 44% 2014 = 41% 2015 = 32% By Field of Membership By Examining Agency 39% 40% 33% 32% 45% 4 43% 32% 47% 44% 42% 33% 28% 27% 35% 28% 49% 47% 40% 31% 5 50% 45% 39% Single Common Bond Multiple Common Bond Community State Regulator NCUA Both By Change in CAMEL Rating By Agreement with CAMEL Rating 38% 35% 29% 1 43% 40% 38% 30% 69% 7 72% 74% 19% 2 20% 11% 39% 44% 39% 32% 6 68% 65% 65% 60% 52% 51% 49% 7 81% 80% 82% Improved Remained the Same Declined Strng Some Neutral Agree Some Agree Strng By Asset Size By Net Worth Ratio 43% 44% 40% 3 49% 4 48% 34% 49% 4 44% 34% 44% 48% 33% 27% 31% 25% 33% 28% 29% 31% 2 11% 65% 70% 77% 20% 78% 74% 100% 58% 50% 50% 44% 34% 38% 37% 37% 31% <$25M $25M- $50M $50M - $250M $250M- $500M $500M-$1B $1B+ < - 7% 7% - 10% 10% > Page 33

DETAILED EVALUATION OF EXAMINERS AND EXAM PROCESS 4 Respondents were asked to rate a variety of issues related primarily to impressions of exam team behaviors and attitudes. The five most positive items evaluated by credit unions, ranked by average level of agreement, is nearly identical to the most positive reported in the 2014 report and the strengh of agreement with each item is little-changed. The most positive items were: 1. The exam team gave management the opportunity to comment or respond prior to sharing results with the board. 2. The exam team was knowledgeable about key safety and soundness issues and regulatory requirements. 3. Examiners were willing to take the time to discuss preliminary exam findings prior to the exit meeting. 4. The exam team was knowledgeable about the credit union being examined. 5. The exam team was flexible and open to discussion with credit union staff. The five most negative items, ranked by average level of agreement, is essentially unchanged compared to the negative items reported in 2014 and the strengh of agreement with each item is similar. The most negative items were: 1. Heavier regulatory/exam requirements are putting increasing pressure on credit union resources. 2. Examiners applied "best practices" as a regulatory standard. 3. Exam team applied "guidance" as if it were enforceable regulation. 4. Examiners are "covering themselves." 5. Items appear in DORs that used to be handled routinely. 4 Average scores are based on a 5-point scale, where 5.0 represents agree strongly and 1.0 represents disagree strongly. Negatively-worded statements are converted to positive to normalize the averages for comparative purposes. Page 34