Proceedings of the Sixteenth Annual Conference of the Applied Business and Entrepreneurship Association International

Proceedings of the Sixteenth Annual Conference of the Applied Business and Entrepreneurship Association International Conference Chair Lisa Andrus Program Co-Chairs Bahram Adrangi Arjun Chatrath Pamplin School of Business Administration The University of Portland November 2019 Kauai, Hawaii, U.S.A Articles published in this Conference Proceedings are accepted based on the double-blind peer-review process Table of Contents Screening Leaders for Success in Turbulent Environments………….5 Use of Alternative Data in Consumer Lending Models: The Case of “Upstart”………………………………………………… …………….10 Child Labor in Globalized Economy: Strategies to Combat the Problem…………………………………………………………… 17 Passively Active Investing – A Five Year Test…………………28 Screening Leaders for Success in Turbulent Environments Phillip L Hunsaker, School of Business, University of San Diego, 5998 Alcala Park, San Diego, CA 92110, Phone: (619 985-8600, philmail@sandiego.edu Abstract The success of task-oriented organizations is highly dependent on the individuals selected to assume responsibility for leadership Because of the high costs involved in leadership training, and the costs related to future consequences, it is important to ensure that individuals who can profit from training and perform successfully in the criterion environment are selected as candidates The purpose of the present study was to test the efficacy of a unique personality variable, the General Incongruity Adaptation Level, as a predictor of success in OCS leadership training The results of the study confirmed that a higher proportion of high GIAL candidates successfully complete the OCS program, which provides support for the basic GIAL hypothesis concerning the relationship between GIAL and environmental turbulence Exposure to the tremendous turbulence in the OCS program resulted in a significant increase of the mean GIAL score of candidates completing the program Low GIAL candidates also reacted more strongly to environmental turbulence than high GIAL candidates, emphasizing the importance of controlling for individual differences when investigating the effects of exposure to incongruent environments Implications for OCS programs of this nature (i.e., producing turbulent-field conditions) include that the GIAL Self- Description Inventory appears to have high potential as a screening device, and that this type of program is instrumental in increasing the adaptation levels of low GIAL candidates Introduction It has been established for some time that the success of task-oriented organizations is highly dependent on the individuals selected to assume responsibility for leadership (Williams, and Leavitt, 1947) Because of the high costs involved in leadership training, and the costs related to future consequences, it is important to ensure that individuals who can profit from training and perform successfully in the criterion environment are selected as candidates Consequently, the determination of effective selection devices is highly desirable This need is especially acute in the Army Officer Candidate School (OCS) where over one-third of the entering class does not graduate, despite an initial screening examination which eliminates approximately 75 percent of all enlisted personnel from OCS consideration (Lippitt and Petersen, 1967), When examining personality characteristics as possible screening criteria, studies have found few significant correlations related to success in OCS leadership training, (Richardson, 1969; Williams and Leavitt, 1947) Although Petersen and Lippitt (1968) found that some OCS candidates have a greater propensity to successfully complete training programs than others, their results were confounded by a variety of design problems making their conclusions only tentative Theoretical Framework, Purpose and Hypotheses The purpose of the present study was to test the efficacy of a unique personality variable as a predictor of success in OCS leadership training The General Incongruity Adaptation Level (GIAL) has been proposed by Driver and Streufert (1965) as an important predictor of responses to turbulent situations (i.e., constantly changing, highly uncertain and ambiguous) Basically, the GIAL is an average expectation of all types of incongruity (e.g., stress, conflict, failure and ambiguity, etc.) Individuals differ in GIAL depending upon their previous experience with incongruity, i.e., the more, incongruity experienced in one's past, the higher his G IAL Environments that provide too little or too much incongruity (i.e., very high or low degree of turbulence) will be disliked, and the individual will attempt to maintain the desired level of environmental turbulence within the range of his GIAL via physical or psychological avoidance, changing the nature of his environment, or the use of other internal defense mechanisms Since the OCS leadership training program is designed to expose candidates to turbulence similar to that encountered in actual combat, they are constantly subjected to mental, physical, and emotional stress (Petersen and Lippitt, 1968) Within this environment, the following relationships with the GIAL concepts were investigated: Hypothesis 1: A greater proportion of high GIAL candidates than low GIAL candidates will successfully complete OCS (Hunsaker, 1975) Hypothesis 2: Experience in OCS will increase candidates' expectations of incongruence Hypothesis 3: The OCS experience will elicit greater increases in the incongruity expectations of low G IA L candidates than high GIAL candidates Hypothesis 4: High GIAL candidates will be more effective leaders than low GIAL candidates in the OCS environment Method Eighty-five cadets of the Wisconsin Army National Guard and Army Reserve completed the GIAL Self-Description Inventory (Driver and Streufert, 1967), immediately prior to, and immediately after, the two-week OCS training program conducted at the Wisconsin Military Academy For comparison, a (nonequivalent) control group consisting of 29 undergraduate students enrolled in the Administrative Organization course at the University of Wisconsin-Milwaukee completed the GIAL inventory on the same dates No significant differences existed between the mean scores of the control group and experimental group on the pre-test administration of the GIAL inventory Comparisons of before and after scores provided evidence of the effects of differences in environmental turbulence on both groups’ GIALs Quartile comparisons provided estimates of the variation of these effects between low and high GIAL subjects Rosters of candidates withdrawing from the training program, and the reasons for these withdrawals, were obtained from the OCS administrative officers The proportions of high GIAL candidates (i.e., scores above the mean) and low GIAL candidates dropping out was determined after eliminating withdrawals due to extraneous reasons such as physical injury Leadership scores, based on observations of the candidates' ability to accomplish assigned missions, were obtained from peer rankings and evaluations by the Tactical Department Officers [Tac officers) who made certain that each candidate was given ample opportunity to exercise leadership skills in turbulent environments Leadership ranks were correlated with GIAL scores to determine the relationship of GIAL level to leadership effectiveness Results The proportion of high GIAL candidates dropping out of the OCS program was 09, while the proportion of low GIAL candidates dropping out was 18 The difference between these proportions was significant (Z = l.76, p 04), resulting in acceptance of the first hypothesis that the proportion of high GIAL candidates successfully completing the program is greater than the proportion of low GIAL candidates completing the program The mean GIAL score of the OCS candidates was 44.87 before exposure to the two-week training program, and 48.12 after completion This 3.25-point difference represents a significant increase (t = 4.12, df = 65, p < 001) in the mean GIAL score The before and after difference between mean GIAL scores for the control group was not significant, and the second hypothesis that subjection to the highly turbulent environment of OCS would result in increases in incongruity expectations was accepted Quartile comparisons revealed significant differences in the changes of incongruity expectations for low and high candidates in OCS, but not in the control group Although OCS candidates in the first (top) quartile and second quartile manifest no significant changes, the mean GIAL scores for candidates in the third quartile increased significantly (t = 2.98, df = 15, p < 01) as did those for candidates in the fourth quartile (t = 6.59, df = 16, p < 001) Because of these differences the third hypothesis that the incongruity expectations of low G IAL candidates would increase by a greater degree than those of high GIAL candidates was accepted Pearson product-moment correlations between GIAL scores and leadership rankings by peers did not yield significant results Correlations between GIAL scores and Tac officers’ leadership rankings also failed to be significant Consequently, the hypotheses suggesting a positive relationship between GIAL scores and leadership in the OCS environment were rejected A significant, negative correlation was found between the leadership rankings of peers and Tac officers (r = 59, Z = 4.72, p < 0001) Since the numerical values in ranking schemes for peers and Tac officers were reversed, the significance of this correlation indicates that both types of judges agreed on candidates' relative leadership capabilities Discussion and Conclusions The positive results confirming the first hypothesis, that a higher proportion" of high GIAL candidates than low GIAL candidates would successfully complete the OCS program, provides support for the basic GIAL hypothesis concerning the relationship between GIAL and environmental turbulence The proposition is that whenever the environment provides either too much or too little turbulence relative to the individual's GIAL, the negative effect associated with this incongruence will motivate the individual to change or avoid it Since an OCS candidate can little to modify the nature of his environment, an active response alternative for overloaded individuals is to withdraw from the program Consequently, low GIAL candidates behave in accordance with traditional dissonance theory and choose to sacrifice the future rewards of becoming an officer in order to avoid the surplus of immediate dissonance relative to their expectations High GIAL candidates, on the other hand, find less discrepancy between this turbulent environment and their expectations Consequently, they have little difficulty enduring the dissonant occurrences and successfully completing the program Support of the second hypothesis suggests an addition to the GIAL model Exposure to the tremendous turbulence in the OCS program resulted in a significant increase of the mean GIAL score of candidates completing the program Thus, when subjected to a situation where they can neither significantly alter the nature of dissonant inputs, nor escape from the situation without considerable cost, it appears that the successful candidates experience at least temporary increases in their incongruity expectations, allowing them to endure the situation, Research is currently in process to determine whether these shifts in expectations arc temporary or permanent The results supporting the third hypothesis that low CIAL candidates react more strongly to environmental turbulence than high GIAL candidates, emphasizes the importance of controlling for individual differences when investigating the effects of exposure to incongruent environments These results also substantiate the GIAL hypothesis that low GIAL individuals will be disturbed by much less turbulence than high GIAL individuals, who may actually seek more incongruence at the same level of environmental turbulence that causes low GIAL individuals to avoid it In terms of the resulting increases in adaptation levels, the largest increase occurred for candidates in the fourth quartile (i.e., lowest CIAL scores), and the second largest for candidates in the third quartile No significant changes occurred for candidates in the top two quartiles (a slight decrease was noted for candidates in the first quartile and a slight increase was noted for candidates in the second quartile) These results suggest that the low GIAL candidates were encountering a degree of environmental incongruity exceeding their adaptation levels, and since withdrawal from the OCS program may have been even more costly (in terms of dissonance experienced) than enduring it, the outcome was an increase in their incongruity expectations High GIAL candidates, on the other hand, may have found the dissonance of OCS training to be congruent with their expectations and, therefore, had no need to adapt Had the level of environmental turbulence been even greater, so that the resulting incongruity exceeded the expectations of both high and low GIAL candidates, the result could have been an increase in the expectations of candidates in all quartiles The lack of significant results regarding the fourth hypothesis indicates that differences in GIAL's are not enough by themselves to predict leadership success rankings in OCS environments Since a significant correlation was found between the leadership rankings of peers and experienced officers, it seems that this is another case, similar to that reported by Williams and Leavitt (1947), where the cadet's fellow candidates are better predictors of leadership effectiveness than personality tests Further research to determine the criteria utilized by these raters, controlling for their own personality make-up, is needed to suggest other personality variables related to leadership success in OCS Implications for OCS programs of this nature (i.e., producing turbulent-field conditions) include the following: (1) the GIAL Self- Description Inventory appears to have high potential as a screening device (2) this type of program is instrumental in increasing the adaptation levels of low GIAL candidates (at least temporarily), (3) although common leadership rankings are produced by peer groups and superior officers, more research is needed to determine the personality and behavioral characteristics contributing to leadership effectiveness References Driver M and S Streufert (1965), The General Incongruity Adaption Level (GIAL) Hypothesis: An Analysis and Integration of Cognitive Approaches to Motivation (W Lafayette, lnd: Purdue University Institute for Research in the Behavioral Economic and Management Sciences Driver M and S Streufert (1967), Purdue-Rutgers Prior Experience Inventory II (GIAL SelfDescription Test, Purdue University Hunsaker, P.L (1975), "Incongruity Adaptation Capability and Risk Performance in Turbulent Decision-Making Environments," Organizational Behavior and Human Performance, Vol 14, No 2, pp 173-185 Hunsaker, P.L (1972) "The Effects of Environmental Incongruity and General Incongruity Adaptation Level on Risk Perception and Risk Preference," Proceedings of the 1972 Annual Convention of the American Psychological Association Hunsaker, P.L., Mudgett, W.C and Wynne, B.E (1975), "Assessing and Developing Administrators for Turbulent Environments," Administration and Society, Vol 17, No 3, pp 312-327 Hunsaker, P.L., Wynne, B.E and Mudgett, W.C (1974), "A Preliminary Model for Developing Managerial Capabilities for Coping with Environmental Turbulence," Proceedings, Midwest Division of the Academy of Management, pp 217-234 Lippitt, G and P Petersen (1967), "Development of a Behavioral Style in Leadership Training." Training and Development Journal, pp 9-17 Petersen, P and G Lippitt (1968), "Comparison of Behavioral Styles Between Entering and Graduating Students in Officer Candidate School." Journal of Applied Psychology, Vol 52, No.1, pp 66-70 Richardson, J (1969), "The Relationship of Some Measures of Candidate Personality and Selection by OTU Board," Australian Military Forces Research Report, Vol 69, pp 1- 26 Tannenbaum, R I., Weschler, R I and F Massarik, Leadership and Organization: A Behavioral Science Approach (New York: McGraw-Hill, 1961) Williams, S., and H Leavitt (1947), "Group Opinion as a Predictor of Military Leadership," Journal of Consulting Psychology, Vol II, pp 283-291 Use of Alternative Data in Consumer Lending Models: The Case of “Upstart” Naveen Gudigantala, Robert B Pamplin School of Business Administration, The University of Portland, 5000 N Willamette Blvd., Portland, OR 97203, Phone: (503) 943-8457 gudigant@up.edu Abstract This work discusses the case of a fin-tech company called Upstart, which specializes in using AI/ML based platform to provide credit to traditionally underserved populations Upstart’s AI platform uses alternative data in addition to the traditional FICO scores in its algorithms This alternative data includes borrowers’ educational data and occupational data Upstart’s data shows that a majority of traditionally underserved populations was able to obtain more credit and at better terms using their credit scoring system Introduction Issues surrounding the fairness of algorithms are attracting much attention from the researchers (Saxena et al., 2019) The goal of this case study is to discuss the opportunities and challenges in using alternative data for credit scoring modeling The case study uses a fin-tech company “Upstart Network, Inc.” (called “Upstart” from here on) and an analysis of Upstart’s AI practices in lending to address the questions of algorithmic fairness in consumer lending In specific, this work will look at how different approaches to the development of machine learning models can either help or hinder fairness in consumer lending This case study is intended for researchers in AI and financial services, students learning analytics/AI, and for practitioners doing AI/Data science work The issues discussed in this case will help students better evaluate the implications of models they learn to create as part of analytics curriculum; for the researchers to continue investigating the problems raised in this study; and for data science practitioners to reflect on issues of algorithmic fairness Consumer Lending and Problems Addressed by Upstart Upstart is an online lending platform, launched by ex-Google employees in 2014, with an aim to provide credit to people with limited credit or work history Consumers in need of credit approach Upstart, a website embedded with Artificial Intelligence (AI)/ Machine Learning 10 ETF Domestic equities Active Active Active Active Indexed MGK VYM VOE VOT VOO International equities Active MSCI EAFE Active MSCI EAFE Active MSCI EAFE Passive/index MSCI/EAFE Emerging markets Emerging markets 28.05% 30,397,316 29,332,144 31,490,361 30,546,352 19,085,678 140,851,85 VCIT BIV VMBS BWX EMAG 15,000,000 5,000,000 5,000,000 5,000,000 5,000,000 35,000,000 -6.05% -7.31% -2.43% -5.38% -11.24% -19.04% 14,091,787 4,634,672 4,878,585 4,731,236 4,437,970 28,336,280 IEFA IXUS VEU VXUS VWO IEMG 18,333,334 18,333,333 18,333,333 10,000,000 17,500,000 17,500,000 100,000,00 17.63% 10.59% 9.38% 9.31% -8.84% -5.86% 21,566,332 20,274,634 20,053,000 10,931,000 15,953,000 16,474,500 105,252,46 25,000,000 25,000,000 21,250,000 21,250,000 21,250,000 21,250,000 25,000,000 25,000,000 25,000,000 20,000,000 230,000,00 6.15% 10.95% 6.08% 2.63% 4.69% -6.69% 38.06% -5.99% 19.78% 0.15% Subtotal Alternative strategies Private equity Private equity Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Venture capital Private real estate Energy and natural resources Distressed debt Subtotal Short-term securities, cash, other Return BDCS BDCL MNA HDG PBP MRGR IWC USRT VAW ANGL 27.99% 23.50% 32.59% 28.62% 27.24% Ending Balance 23,750,000 23,750,000 23,750,000 23,750,000 15,000,000 110,000,00 Subtotal Fixed income Domestic investment grade - active Domestic investment grade - passive Domestic non-investment grade International bonds Emerging markets Subtotal Beginning Balance 5.25% 8.13% 26,536,643 27,736,411 22,540,967 21,809,348 22,247,099 19,828,067 34,515,714 23,503,586 29,944,896 20,029,630 248,692,36 Short-term securities, cash Total VGSH 25,000,000 0.05% 25,012,325 500,000,00 9.63% 548,145,28 Taking the ending balance from 2013 and using it as the beginning balance in 2014, and then applying the average asset allocation to the beginning balance and applying the returns produced a portfolio of: ETF Domestic equities Active Active Active Active Indexed MGK VYM VOE VOT VOO Subtotal Beginning Balance Return 12.50% 10.66% 12.41% 12.91% 11.79% Ending Balance 24,666,538 24,666,538 24,666,538 24,666,538 27,407,264 126,073,41 12.05% 27,749,855 27,294,777 27,726,602 27,851,616 30,638,631 141,261,48 Fixed income Domestic investment grade - active Domestic investment grade - passive International bonds Subtotal VCIT BIV BWX 27,407,264 10,962,906 5,481,453 43,851,623 4.04% 3.57% -3.66% 2.96% 28,513,357 11,354,438 5,281,052 45,148,847 International equities Active MSCI EAFE Active MSCI EAFE Passive/index MSCI/EAFE Emerging markets Emerging markets Subtotal IEFA IXUS VXUS VWO IEMG 21,925,811 21,925,811 21,925,811 16,444,358 16,444,358 98,666,151 -7.89% -6.63% -6.73% -1.21% -3.90% -5.57% 20,195,403 20,473,058 20,449,425 16,245,451 15,803,028 93,166,366 BDCS BDCL MNA HDG PBP MRGR IWC IPO USRT VAW ANGL 27,407,264 27,407,264 24,666,538 24,666,538 24,666,538 24,666,538 16,444,358 16,444,358 27,407,264 27,407,264 16,444,358 257,628,28 -15.20% -28.21% 5.72% 1.83% 0.14% -2.08% 2.52% 4.20% 25.12% 4.15% -3.59% 23,240,791 19,676,304 26,077,389 25,117,936 24,702,286 24,154,315 16,858,260 17,135,021 34,291,725 28,544,914 15,853,798 255,652,73 Alternative strategies Private equity Private equity Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Venture capital Venture capital Private real estate Energy and natural resources Distressed debt Subtotal Short-term securities, cash, other -0.77% Short-term securities, cash Total VGSH 21,925,811 0.02% 21,929,413 548,145,28 1.64% 557,158,84 procedure for 2015: Beginning Balance EFT Domestic equities Active Active Active Active Indexed MGK VYM VOE VOT VOO International equities Active MSCI EAFE Active MSCI EAFE Passive/index MSCI/EAFE Emerging markets Subtotal Alternative strategies Private equity Private equity Marketable alternative strategies Marketable alternative strategies Marketable -1.93% 25,492,729 24,229,997 23,946,562 24,492,445 27,510,619 125,672,35 VCIT 27,857,942 -2.47% 27,169,733 BIV 16,714,765 44,572,708 -1.91% -2.26% 16,394,998 43,564,731 IEFA 27,857,942 -2.16% 27,256,475 IXUS 27,857,942 -7.38% 25,803,276 VXUS 11,143,177 -6.68% 10,398,608 VWO 27,857,942 94,717,004 -18.12% -8.92% 22,810,083 86,268,442 BDCS BDCL 25,072,148 25,072,148 -11.44% -26.49% 22,203,632 18,429,297 MNA 27,857,942 -0.04% 27,848,021 HDG PBP 27,857,942 27,857,942 -0.05% -0.82% 27,844,711 27,628,936 12 1.68% -3.36% -4.49% -2.31% -1.25% Ending Balance 25,072,148 25,072,148 25,072,148 25,072,148 27,857,942 128,146,53 Subtotal Fixed income Domestic investment grade - active Domestic investment grade - passive Subtotal Return alternative strategies Marketable alternative strategies Venture capital Venture capital Private real estate Energy and natural resources Commodities and managed futures Distressed debt MRGR IWC IPO 27,857,942 19,500,560 19,500,560 -0.55% -6.64% -0.21% 27,705,713 18,205,236 19,459,445 USRT 27,857,942 -12.54% 24,363,413 VAW 16,714,765 -37.33% 10,474,908 GSP 5,571,588 -37.33% 3,491,714 ANGL 16,714,765 267,436,24 -6.31% -9.02% 15,660,064 243,315,09 22,286,354 -0.23% 22,235,121 -6.48% 521,055,73 Subtotal Short-term securities, cash, other Short-term securities, cash VGSH 557,158,84 Total 12 And for 2016: EFT Domestic equities Active Active Active Active Indexed MGK VYM VOE VOT VOO International equities Active MSCI EAFE Active MSCI EAFE Passive/index MSCI/EAFE Emerging markets Return 11.23% 23,651,784 25,511,885 25,395,396 23,836,694 29,106,479 127,502,23 VCIT BIV VMBS 20,842,229 5,210,557 5,210,557 31,263,344 1.91% -0.18% -0.98% 1.08% 21,241,278 5,201,166 5,159,494 31,601,937 IEFA IXUS VXUS VWO 23,447,508 23,447,508 26,052,787 15,631,672 0.30% 4.21% 3.15% 12.37% 23,517,671 24,434,772 26,872,794 17,565,993 12 6.80% 15.20% 14.68% 7.64% 11.72% Ending Balance 22,144,869 22,144,869 22,144,869 22,144,869 26,052,787 114,632,26 Subtotal Fixed income Domestic investment grade - active Domestic investment grade - passive Domestic non-investment grade Subtotal Beginning Balance Emerging markets IEMG Subtotal Alternative strategies Private equity Private equity Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Venture capital Venture capital Private real estate Energy and natural resources Distressed debt BDCS BDCL MNA HDG PBP MRGR IWC IPO USRT VAW ANGL Subtotal Short-term securities, cash, other Short-term securities, cash VGSH Total 12 15,631,672 104,211,14 10.49% 23,447,508 23,447,508 24,750,147 24,750,147 24,750,147 24,750,147 23,447,508 23,447,508 20,842,229 20,842,229 20,842,229 255,317,31 13.85% 25.48% 4.98% 2.70% 4.62% -2.38% 20.73% 0.34% 4.63% 21.20% 18.44% 10.12% 26,694,629 29,421,396 25,982,778 25,419,230 25,894,522 24,160,536 28,308,496 23,527,184 21,808,100 25,261,402 24,684,747 281,163,02 15,631,672 0.21% 15,665,101 521,055,73 8.55% 565,594,96 5.23% 17,271,434 109,662,66 And finally for 2017: EFT Domestic equities Active Active Active Active Indexed MGK VYM VOE VOT VOO International equities Active MSCI EAFE Active MSCI EAFE Active MSCI EAFE Passive/index MSCI/EAFE Emerging markets 18.36% 32,361,191 28,590,258 28,946,527 30,523,009 26,850,162 147,271,14 VCIT BIV 28,279,748 11,311,899 39,591,647 2.27% 1.23% 1.97% 28,921,793 11,451,232 40,373,025 IEFA IXUS VEU VXUS VWO 26,394,432 26,394,432 26,394,432 11,311,899 28,279,748 118,774,94 22.71% 24.34% 23.05% 23.10% 26.89% 32,387,820 32,820,042 32,478,149 13,924,789 35,885,109 147,495,90 Subtotal Alternative strategies Private equity Private equity Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Marketable alternative strategies Venture capital Venture capital Private real estate Energy and natural resources Commodities and managed futures Distressed debt Return BDCS BDCL MNA HDG PBP MRGR IWC IPO USRT VAW GSP ANGL Subtotal 12 25,451,773 25,451,773 25,451,773 25,451,773 25,451,773 25,451,773 22,623,798 22,623,798 22,623,798 22,623,798 5,655,950 22,623,798 271,485,58 27.15% 12.33% 13.73% 19.92% 18.68% Ending Balance 25,451,773 25,451,773 25,451,773 25,451,773 22,623,798 124,430,89 Subtotal Fixed income Domestic investment grade - active Domestic investment grade - passive Subtotal Beginning Balance 24.18% -8.79% -16.50% 5.65% 5.15% 0.56% 1.71% 10.19% 35.49% 1.99% 20.90% 5.79% 3.85% 5.01% 23,215,238 21,252,877 26,889,335 26,762,442 25,594,962 25,887,273 24,928,894 30,654,052 23,074,972 27,351,846 5,983,217 23,494,548 285,089,65 Short-term securities, cash, other Short-term securities, cash VGSH 11,311,899 565,594,96 Total 12 -0.63% 11,241,037 11.65% 631,470,77 Analysis The returns from the Studies versus the returns from the replacement portfolio were: Year Study Replacement Portfolio 2013 2014 2015 2016 2017 11.9% 7.1% 1.1% 6.7% 14.3% 9.6% 1.6% -6.5% 8.6% 11.7% 5-year return 8.1% 4.8% The actively managed portfolios from the various Studies outperformed the replacement portfolio is four of the five years, and the cumulative recalculated 5-year return was also superior The returns can also be compared based on the asset class allocation: US Equities Year Study Replacement Portfolio 2013 2014 2015 2016 2017 5- year 31.8% 9.9% -1.3% 10.2% 21.5% 13.9% 28.1% 12.2% -1.9% 11.2% 18.4% 13.1% 12 Year 2013 2014 2015 2016 2017 5- year Year 2013 2014 2015 2016 2017 5- year Year 2013 2014 2015 2016 2017 5- year Fixed Income Replacement Study Portfolio -0.7% 4.2% 0.1% 2.9% 3.8% 2.0% -19.0% 3.0% -2.3% 1.1% 2.0% -3.4% Non-US Equities Replacement Study Portfolio 15.9% 0.2% -5.0% 4.6% 28.2% 8.1% 5.3% -5.6% -8.9% 5.2% 24.2% 3.4% Alternative Strategies Replacement Study Portfolio 7.3% 14.2% -2.1% 7.1% 9.8% 7.1% 8.1% -0.8% -9.0% 10.1% 5.0% 2.5% 12 Year 2013 2014 2015 2016 2017 5- year Short-term Securities/Cash Replacement Study Portfolio 0.1% 0.4% 0.1% 1.2% 0.8% 0.5% 0.5% 0.2% -0.2% 0.2% -0.6% 0.0% The returns in US Equities and Short-term Securities/Cash was fairly close The larger differentials were in Fixed Income (5.4%), Non-US Equities (4.7%) and Alternative Strategies (4.6%) It should not be overlooked that the replacement portfolio was an entirely naïve portfolio – all ETFs selected for the portfolio were done without reference to past returns, and once selected the ETF remained in the portfolio unless a future asset allocation required the removal of an ETF in a particular allocation The ETF to be removed was the last one listed, not the worst performer It should also be pointed out the ETFs that were selected were based on a higher Morningstar rating and/or being designated an “all-star,” so the extent to which this represents a better performing ETF then the replacement portfolio may not have been so naïve The take-away is that from the data in this study, passive investing produces comparable returns to active investing for allocations to US Equities and Short-term Securities/Cash, and active investing outperforms passive for allocations to Fixed Income, Non-US Equities and Alternative Strategies 12 References Council on Foundations-Commonfund, “2013 Study of Investments for Private Foundations.” Council on Foundations-Commonfund, “2014 Study of Investment of Endowments for Private and Community Foundations.” Council on Foundations-Commonfund, “2015 Study of Investment of Endowments for Private and Community Foundations.” Council on Foundations-Commonfund, “2016 Study of Investment of Endowments for Private and Community Foundations.” Council on Foundations-Commonfund, “2017 Study of Investment of Endowments for Private and Community Foundations.” 12 12 ... response to the same problem, as an joined initiative of UNICEF, Save the Children, the World Federation of the Sporting Goods Industry (WFSGI), the Soccer Industry of America and the ILO (and others)... expectations High GIAL candidates, on the other hand, may have found the dissonance of OCS training to be congruent with their expectations and, therefore, had no need to adapt Had the level of environmental... that the resulting incongruity exceeded the expectations of both high and low GIAL candidates, the result could have been an increase in the expectations of candidates in all quartiles The lack of