Economic Effects on Million Dollar Giving

Abstract

This study investigates the impact of economic conditions on the number of charitable gifts of 1 million dollars or more within the United States using the Million Dollar List (MDL) data set. We investigate key donor types—individuals, corporations, and foundations—using quarterly data. Results indicate that individual donors are significantly responsive to underlying economic conditions, foundation giving tends to be countercyclical, and corporate giving is less closely linked with aggregate macroeconomic conditions. We also find that economic conditions vary in their influence on million dollar giving to subsectors, and gifts to public benefit and human services organizations increase significantly during periods of recession, holding other factors constant. In contrast, million dollar giving to arts and education organizations is significantly associated with favorable economic conditions, holding other factors constant. Findings have direct implications for philanthropists, fundraisers, and policy makers as they seek to understand how economic conditions affect large gifts.

Keywords

philanthropy economy giving high net worth giving recession

Introduction

Million dollar gifts account for a significant share of U.S. charitable dollars; as an example, the 2016 Million Dollar Donors Report found that 1,823 million dollar gifts were given within the United States in 2015, with the total value of the donations at US$19.3 billion (Coutts, 2016). The total value of these gifts constitutes nearly 5% of the US$390.05 billion total estimated for charitable giving during 2016 (Giving USA, 2017). Recent policy debates on growing wealth and income inequality in the United States have led to a great deal of interest in the charitable behavior of wealthy Americans (Piketty & Saez, 2006). In general, charitable giving has been shown to be highly skewed, with several studies showing that high net worth donors contribute a disproportionately large share of all philanthropic dollars (Havens, O’Herlihy, & Schervish, 2006; Schervish & Havens, 2001). Scholars have also emphasized that understanding high net worth giving is an important component of philanthropy and the nonprofit sector.

Recent estimates suggest that U.S. charitable giving has grown substantially between 1975 and 2015, with an average year-to-year inflation-adjusted dollar increase of US$6.19 billion (Giving USA, 2016). Overall, inflation-adjusted charitable giving makes up 2% of gross domestic product (GDP; Giving USA, 2016). Given the recent economic downturn, there has been a great deal of interest in understanding how charitable giving and its composition react to changing economic conditions. Between 2007 and 2009, total charitable giving declined by 9.75%, with a 6.5% decline taking place between 2008 and 2009 alone. Research has documented how economic conditions have influenced charitable giving by individuals, foundations, and corporations, as well as how changes have influenced giving to various subsectors, such as higher education, health, arts, and basic needs (Congressional Budget Office, 2011; Drezner, 2006; Galaskiewicz, 1997; List & Peysakhovich, 2011). Because large gifts have the power to influence and shape trends in philanthropy, understanding million dollar giving over the business cycle is an important aspect of philanthropy and the nonprofit sector, yet it remains underexplored.

This article investigates how economic conditions have influenced large gifts using a unique data set on million-dollar-plus gifts. Few studies have sought to disentangle the nature of this relationship with regard to specific types and levels of contributions. To our knowledge, this article is one of the first to establish empirical evidence for a link between economic conditions and million dollar giving, using the Million Dollar List (MDL). The MDL is a publicly available data set of publicly announced gifts valued at US$1 million or greater originating in the United States since 2000 (Indiana University Lilly Family School of Philanthropy, 2012). This article is based on data from the years 2000 to 2011. Data collection involves manual coding of gift announcements from a variety of publications, including the Chronicle of Philanthropy, Google email alerts, and the FoundationSearch database. The MDL is compiled by the Indiana University Lilly Family School of Philanthropy and is available at https://milliondollarlist.org/.

Existing research has examined the relationship between stock market performance and overall charitable giving. List and Peysakhovich (2011) find that giving is more likely to respond positively to increases in Standard & Poor’s 500 index (S&P 500) than it is to respond negatively to decreases in the S&P 500. In fact, the S&P 500 accounts for 40% of the variation in the percentage change of total annual charitable giving. One particularly interesting finding is that changes in the S&P 500 are shown to affect giving to charitable subsectors differently. For example, changes in giving to religion are not significantly associated with changes in the S&P 500, whereas giving for educational purposes follows changes in the S&P 500 more closely. Figure 1 provides some initial evidence that economic factors influence overall trends in gifts at the million dollar level on the donor end of the equation. Measured on a quarterly basis, gifts from individuals appear to be highly correlated with the S&P 500, gifts from foundations tend to increase as the S&P 500 drops, and gifts from corporations appear unaffected by movement in the index.

Figure 1.

Million-dollar-plus gifts by donor type and the S&P 500.

This article investigates how million-dollar-plus giving (measured by the number of million dollar gifts) is affected by economic conditions (measured by inflation-adjusted values of the S&P 500 index, the quarterly unemployment rate, Personal Consumption Expenditures [PCE], and U.S. GDP). This research uses a unique data set which measures publicly announced giving on a quarterly basis and seeks to address two primary questions. First, we investigate how economic trends affect million-dollar-plus giving by analyzing this giving across three broad donor types: individuals/households, foundations, and corporations. Second, we examine million dollar gifts to various nonprofit subsectors to discern the impact of economic conditions on specific subsectors. Ultimately, an analysis of these donors and subsectors can inform a practitioner’s understanding of how to effectively mobilize large gifts in times of economic volatility. In the aftermath of the Great Recession, the results in this article are of increasing importance because policy makers and nonprofit leaders are often interested in the impact of economic conditions on giving patterns by donor type and subsector, and research that delves into these questions is limited.

We find significant differences between the three broad donor categories (individuals, corporations, and foundations) with regard to the impact of macroeconomic conditions on publicly announced million-dollar-plus gifts. Specifically, giving by individuals is significantly influenced by contemporaneous economic trends. The frequency of million-dollar-plus gifts by individuals is diminished during adverse economic conditions. Foundation giving appears to be countercyclical, with many foundations increasing giving, particularly to human services organizations, during times of recession. Interestingly, million dollar giving by corporations appears to be relatively unaffected by macroeconomic conditions and tends to be affected by industry-specific conditions. When we examine giving by subsector, million dollar giving to arts and higher education organizations appears to be negatively influenced by adverse economic indicators compared with health and human services.

Conceptual Framework: Giving by Households, Corporations, and Foundations

Individuals comprise a key component of overall charitable giving and make up nearly 75% of charitable giving annually (Giving USA, 2016). Researchers have become increasingly interested in understanding how household giving is influenced by economic factors including income, wealth, and tax policies. One testable hypothesis is that economic conditions will tend to have a larger impact on giving by individuals and households compared with their effect on foundations and corporations, given that individuals do not face the same institutional constraints that may influence corporate and foundation giving, such as payout requirements on foundations.

Recent studies have also emphasized that noneconomic motivations for individual charitable giving, such as desire for status, prestige, and social pressure by fundraisers, as well as altruism and “warm glow,” may influence overall patterns of individual giving (Andreoni, 1993; Andreoni & Petrie, 2004; DellaVigna, List, & Malmendier, 2009; Kingma, 1989; Long, 1976; Samek & Sheremeta, 2014). According to another model of altruism, impact philanthropy, an individual donor might be motivated by the increased need that an economic downturn creates and may respond by giving more (Duncan, 2004). This model suggests that household giving may shift across organization types and subsectors if donors perceive that their charitable contributions are more valuable to an organization or cause.

Foundation giving has made up about 15% of all charitable giving in recent years (Giving USA, 2016). Existing research appears to suggest that macroeconomic conditions have less of an effect on foundation donors than on individual donors. Indeed, in a recent study there was no significant correlation between private foundation giving and macroeconomic or institutional variables (Lew & Wójcik, 2009). Several factors may contribute to foundation giving being less responsive to macroeconomic conditions. Internal factors, such as using rolling averages to devise grantmaking budgets and priorities and using reserves to make planned gifts, may outweigh economic factors in influencing foundation charitable giving (Drezner, 2010; Foundation Center, 2010). Foundations tend to be more proactive and dedicated to selecting program areas in which to invest and increase their giving during adverse economic periods (Katz, 2005). Indeed, some foundations may emphasize meeting social needs as a primary area of grantmaking (Lew & Wójcik, 2009; Reich, Wimer, Mohamed, & Jambulapati, 2011).

Corporate charitable giving typically makes up about 5% of all charitable dollars in any given year (Giving USA, 2016). A significant body of literature exists on the extent to which economic factors affect corporate giving. Much of the literature examines whether corporations engage in philanthropic activity to advance their profit goals versus corporate social responsibility objectives (Moir & Taffler, 2004; Reich et al., 2011). Industry structure, state and local policies, size, and corporate profits for a given firm have been shown to have strong relationships with corporate charitable giving (Drezner, 2010; Giving USA Foundation, 2009; Kitzmueller & Shimshack, 2012; Seifert, Morris, & Bartkus, 2003). Corporate giving can also be influenced by firm and industry performance, as well as business needs (Moir & Taffler, 2004; Urriolagoitia & Vernis, 2012). Another testable hypothesis is that corporate giving may be less closely linked to overall economic conditions than individual giving. Although economic downturns have an aggregate negative effect on corporate giving, this tends to depend on subsector or industry type rather than macroeconomic indicators (Amato & Amato, 2007; Galaskiewicz, 1997).

Data

This article uses the MDL, a publicly available data set providing an in-depth view of high net worth giving. The MDL provides a comprehensive picture of publicly announced gifts valued at US$1 million or greater originating in the United States since 2000. The MDL is compiled by the Indiana University Lilly Family School of Philanthropy (formerly the Center on Philanthropy). Providing a unique perspective, the MDL has developed in-depth information on giving by American individuals, corporations, foundations, and other grantmaking nonprofit organizations. The current project is based on data from the years 2000 to 2011.

The main advantage of the MDL is that it provides gift-level information on a quarterly basis, allowing for better understanding of how overall economic conditions affect giving trends and patterns. This gift-level view contrasts with data sources that have been used in previous studies on charitable giving. These other sources include tax data, which are only available in an aggregate form: Statistics of Income (SOI) data provide an organization-level view and do not disaggregate the information according to gift level, and individual tax level data only provide an aggregate amount and are not broken out by specific gifts or subsectors (Internal Revenue Service, 2016). Another popular resource, the Philanthropy 400, looks at individual aggregate giving (Olsen-Phillips, 2016). Furthermore, the MDL data are available on a quarterly basis, in contrast to other data sources that are only published annually. However, the methodology does have some limitations in its reliance on public announcements surrounding million-dollar-plus gifts. Public announcements through media sources may vary based upon the volume of gifts made at a certain time or various factors surrounding the gift, which opens the possibility of bias in the data. Public announcements may reflect overall trends in giving, because the timing of these announcements may correspond with generosity for wealthy individuals, fulfillment of corporate social responsibility promises by businesses, or fundraising strategies by foundations. This in mind, our model remains indicative of overall trends as it does capture any large gift that has a public announcement. Furthermore, we examine pairwise correlations between annual Internal Revenue Service (IRS) data and our annual MDL data and find strong relationships.

Although most studies on U.S. charitable giving by individuals, corporations, and foundations rely on household surveys and tax records, both have clear strengths and limitations. The primary limitation of data based on tax return sampling is that this information typically only includes aggregate donations reported on an individual’s tax return, which may not include all of their charitable giving (Joulfaian, 2005). In addition, tax data tend to provide aggregate information and are not typically publicly available at the gift level (Rooney, Steinberg, & Schervish, 2001), and analysis based on tax data is restricted to the tax effects on itemized charitable giving (Wilhelm, 2006). Yet little evidence is provided about specific gifts, as the names of the organizations that receive these gifts may not be publicly available. The primary limitation of household surveys on charitable giving is that very few household data sources are available specifically on million dollar giving, making it extremely difficult to estimate differences at the highest levels of giving distributions (Wilhelm, 2007). With the lack of gift-specific data, and most data sources providing only annual household giving levels, it is hard to measure the impact of economic conditions specifically on large gifts.

The MDL’s data collection sources include the following: the Chronicle of Philanthropy’s monthly publication and attendant website, the Chronicle of Higher Education’s weekly publication, NOZA Search’s weekly announced gifts, Factiva, LexisNexis Academic, the Philanthropy News Digest from the Foundation Center, Google News email alerts, and the FoundationSearch database. Many of these sources provide daily and weekly updates. Once qualifying gifts are identified, researchers code each gift and enter it into a central database. Specific data that are coded for each gift include donor name, city, state, and type; recipient name, state, country, and subsector; gift amount and notes; source of information; date reported; and year and quarter of the donation. Giving USA estimates that the MDL captures 25% of all gifts at this level in the United States (Giving USA, 2014).

We note that there may be some variation in how data from different subsectors are represented on the MDL. For example, gifts to higher education may be overrepresented and gifts to religious organizations may be underrepresented, due to the different preferences of these types of organizations for publicity and media attention. In the sample used for this analysis, we primarily rely on publicly announced data, as tax data do not offer quarterly updates to reflect the specific time period of giving. Gifts made by individuals through bequest (1.9% of the number of gifts in the sample, 15% of the number of gifts from individuals) are excluded, as the decision making governing bequests may differ from individual gifts; the timing of a bequest generally does not depend on economic factors. For example, bequests may be accidental, as a consequence of uncertain lifespan and an imperfect market for life annuities (Davies, 1981; Hurd, 1989). Gifts marked as being given by “other” groups (7.5% of the number of gifts in the sample) are also excluded. We focus the analysis on three major donor types: individuals, foundations, and corporations and corporate foundations. These groups account for approximately 91% of the total number of gifts that reflect quarterly changes and 83% of corresponding dollar values. All dollar figures are inflation-adjusted to 2011 values. Note that 2011 values are used for this section only; all regressions are inflation-adjusted to 2005 dollars.

There are some notable differences in the giving trends among the three primary donor types on the MDL (Table 1). Among donor types, we note the number of gifts varies by donor type, and the percentage of total value that those gifts represent also tends to vary. Interestingly, individuals and households give fewer gifts than foundations and more than corporations, but when comparing dollar amounts, individuals and households donate the most, representing 60% of the total value. The results of our analysis illustrate the striking difference in the value of gifts contributed by individual donors as opposed to foundations and corporations. Among the subsectors receiving the million dollar gifts, higher education receives a significantly higher proportion, both of the number of gifts (47%) and value of the gifts (34%) (Table 2). In the empirical analysis, after considering multiple alternate specifications, we focus on the number of gifts rather than dollar value of gifts because the distribution of large gifts tends to be highly skewed. However, we consider several robustness checks with the dollar value of gifts as the key dependent variable.

Table 1.

Million-Dollar-Plus Giving by Donor Type, 2000-2011.

	All donors	Individuals	Corporations	Foundations
Total dollar	226,000	137,000 (60.62%)	20,300 (8.98%)	68,200 (30.18%)
Mean dollar	11.80	19.30	6.77	7.60
Median dollar	2.38	2.61	1.70	2.44
N	19,080	7,092 (37.17%)	3,004 (15.74%)	8,984 (47.09%)

Note. Donors only include individuals, corporations, and foundations. Dollar amount is in millions and inflation-adjusted to 2011 values.

Table 2.

Million-Dollar-Plus Giving by Recipient Subsector, 2000-2011.

	Human services and public/societal benefit	Arts, culture, humanities	Health	Higher education	Other subsectors
Total dollar	15,800 (6.99%)	12,000 (5.31%)	13,900 (6.15%)	76,400 (33.79%)	108,000 (47.77%)
Mean dollar	4.95	7.94	9.19	8.48	N/A
Median dollar	1.85	2.10	2.58	2.54	N/A
N	3,198 (16.76%)	1,509 (7.91%)	1,511 (7.92%)	9,011 (47.23%)	3,851 (20.18%)

Note. Dollar amount is in millions and inflation-adjusted to 2011 values.

Based on the initial analysis of the MDL, a number of patterns have emerged. First, the data collection may underreport gifts made to religious organizations and small nonprofits, both of which are less likely to publicly report or obtain media coverage of such gifts. Second, specific gifts as reported may differ from the actual size of the gift or estimated value, for instance, of nonmonetary contributions such as artwork, stock, or in-kind support. Finally, there may be some duplication in gift reporting due to variation in how the media covers these contributions and the timing of the reports. However, we have examined several specification checks to deal with this concern, including examining the number of donors as a dependent variable.

To study the impact of macroeconomic conditions on million-dollar-plus giving, we use aggregate economic data drawn from various sources. The S&P 500 index is based on the closing price on the last day of the quarter (Yahoo! Finance, 2012). The unemployment rate is estimated by the Bureau of Labor Statistics (U.S. Bureau of Labor Statistics, 2012). GDP and PCE are seasonally adjusted at an annual rate (U.S. Bureau of Economic Analysis, 2012). Recession is included as a dummy variable for whether any months in a quarter were recessionary (NBER, 2012). See Table 3 for more detailed definitions of the independent variables utilized for this study.

Table 3.

Independent Variable Descriptions.

Independent variable	Definition	Source	M	SD	Avg. quarterly % change	Abs. Avg. quarterly % change
Standard & Poor’s 500 Index (S&P 500)	The index uses the closing price on the last day of the quarter. It represents the changes in stock market based on the performance of the 500 largest capitalization stocks trading in the United States.	Yahoo! Finance	1,178.871	208.33	−0.50	7.0
GDP	GDP measures the value of final goods and services produced in the United States in a given period of time. The GDP measure is in 2005 dollar value.	U.S. Bureau of Economic Analysis	12,436.35	755.32	0.40	0.7
PCE index	PCE includes goods and services purchased by individuals, the operating expenses of nonprofit institutions serving individuals, the value of food, fuel, clothing, rent of dwellings, the financial services received in kind by individuals, and net purchases of used goods.	U.S. Bureau of Economic Analysis	98.640	6.927	0.50	0.7
Unemployment rate	Unemployment rate is the estimated percentage of unemployed people aged 16 years and older.	U.S. Bureau of Labor Statistics	6.194	1.941	1.8	4.3
Recession	Recession is identified by the NBER definition of significant declines in economic activities, see: http://www.nber.org/cycles/recessions.html	NBER	0.229	0.425

Note. Dollar amount of MDL gifts is converted into 2005 dollar value for regression purposes. GDP = gross domestic product; PCE = Personal Consumption Expenditures; NBER = National Bureau of Economic Research; MDL = Million Dollar List.

Method

To analyze the effects of economic variables on giving, we classify the MDL data by donor type and subsector. The key dependent variable of interest is the number of million dollar gifts by donor type and to a given subsector in each quarter (analysis by number of donors and dollar amount is provided as a robustness check). By studying the incidence (number) of gifts, rather than the dollar amount, we reduce the role of measurement error in our analysis. We have also examined an alternative dependent variable, measured as the level of donations measured in logs.

For each donor type and subsector, the linear model

N u m b e r o f G i f t s_{d s, t} = α + β I n d i c a t o r_{t} + γ Y e a r_{t} + ζ Q u a r t e r_{t} + ε_{d s, t}

was used to estimate the impact of each economic indicator on the number of gifts received per quarter, where ds represents a specific donor type to a specific subsector, Year represents a vector of year dummy variables, and Quarter represents a vector of quarter dummy variables. We assess the size and significance of the estimator for the economic indicator variable (β) to determine whether it is associated with the number of gifts over the time period. We analyze the impact of each economic indicator on the number of gifts received per quarter for three different time periods: during the same quarter, and lagged both one and four quarters. We include the following economic indicators as independent variables in our analysis: S&P 500 index, GDP, PCE index, unemployment rate, and recession (defined in Table 3). Quarter and year are also included as independent variables.

The baseline model in our analysis is an OLS (ordinary least squares) regression which tests the impact of economic conditions on million-dollar-plus giving. We deal with potential problems associated with multicollinearity (where two or more variables are highly correlated) by using separate panels for each economic indicator. We have considered alternative specifications conducted as a series of robustness checks. These results are available upon request and include Poisson regression models, a variety of dependent variables, and specifications that include interactions between variables.

Results

Economic Effects on Giving by Donor Type

Table 4 presents the baseline model. In this model, the dependent variable is the number of million dollar gifts by donor type in each quarter. We include the following independent variables in our analysis: S&P 500 index, GDP, PCE index, unemployment rate, and recession. The S&P 500 index is based on the closing price on the last day of the quarter (Yahoo! Finance, 2012). The unemployment rate is estimated by the Bureau of Labor Statistics (U.S. Bureau of Labor Statistics, 2012). GDP and PCE are seasonally adjusted at an annual rate (U.S. Bureau of Economic Analysis, 2012). Recession is included as a dummy variable for whether any months in a quarter were recessionary (NBER, 2012). See Table 3 for more detailed definitions of the independent variables utilized for this study.

Table 4.

Impact of Economic Indicators on Incidence of Million-Dollar-Plus Giving in the Same Quarter.

Dependent variable: Number of million-dollar-plus gifts per quarter by each donor type
	All donors	Individuals	Corporations	Foundations
Panel A: S&P 500
β: S&P 500	0.0611	0.1733***	−0.0279	−0.0843
	(0.0639)	(0.0385)	(0.0354)	(0.0541)
	[0.1181]	[0.6992]	[−0.1744]	[−0.3432]
Panel B: GDP
β: GDP	0.1836**	0.1792***	−0.0084	0.0128
	(0.0871)	(0.0468)	(0.0376)	(0.0774)
	[1.3028]	[2.6197]	[−0.1931]	[0.1064]
Panel C: PCE
β: PCE	16.6977*	18.5623***	5.0141	−6.8788
	(9.4926)	(6.4946)	(4.3684)	(8.3701)
	[1.0653]	[2.5177]	[0.9410]	[−1.0516]
Panel D: Unemployment rate
β: Unemployment Rate	−39.1374**	−42.3727***	11.5680	−8.3327
	(16.7717)	(12.9871)	(12.4406)	(18.4559)
	[−0.7063]	[−1.5906]	[0.6390]	[−0.2582]
Panel E: Recession
β: Recession	54.6049	−9.0370	9.0000	54.6420***
	(34.1203)	(21.1247)	(6.6027)	(12.8455)
	[0.2169]	[−0.0817]	[0.1138]	[0.4341]
Number of donations	19,077	7,092	3,003	8,982
N (Collapsed to quarters)	48	48	48	48

Note. Each cell shows a separate ordinary least squares (OLS) regression with a different independent variable as an indicator economic health (β), reporting the variable’s coefficient from a regression: Number of Gifts_d,_t = α + βIndicator_t + γYear_t + ζQuarter_t + ε_ds,t. Robust standard errors in parentheses. Standardized beta coefficients in brackets. S&P = Standard & Poor’s; GDP = gross domestic product; PCE = Personal Consumption Expenditures.

Significant at the 10% level. **Significant at the 5% level. ***Significant at the 1% level.

Giving by individuals

We begin our analysis by studying million dollar giving by individuals. In general, the results in Table 4 show that individual donors at the million dollar level tend to be responsive to overall economic conditions. In our analysis, we discuss the economic indicators that are closely associated with individual giving.

First, we analyze the relationship between the stock market and individual million dollar giving, as this relationship has been explored in the existing literature. We find that a 100-point increase in the value of the S&P 500 index in a given quarter is associated with an increase of more than 17 gifts from individual donors over the same quarter.

The PCE also significantly influences the number of individual gifts in a given quarter: a 1-point increase in the PCE index is associated with an increase of 19 gifts in the same quarter. This represents an economically significant change in individual million dollar giving, as the average number of individual gifts within a quarter is approximately 148.

Importantly, million dollar giving by individuals is negatively associated with unemployment. This result further supports the finding that million dollar giving by individuals tends to be influenced by the business cycle. A 1 percentage point increase in the unemployment rate is associated with a 42-gift decrease by individuals in the same quarter.

Additional findings support individual giving at the million dollar level tends to be procyclical; changes in GDP are also positively associated with individual million dollar donations. Specifically, an increase of US$100 billion in the GDP is associated with an 18-gift increase in giving by individuals during the same quarter. We also find that recession, though not statistically significant, is negatively related to the number of individual gifts: When the United States is in a recession in a particular quarter, million dollar giving by individuals tends to fall by nine gifts in the same quarter.

Giving by foundations

In general, million dollar gifts by foundations appear to be less closely associated with overall economic conditions compared with gifts by individuals. Interestingly, we find that only the recession variable is significantly positively associated with giving by foundations in the same quarter, making foundation giving countercyclical. During a quarter in which the United States was in economic recession, foundations were shown to give 55 more million dollar gifts and grants on average. This is a substantial change in foundation giving, considering that the average number of such gifts and grants in any given quarter is about 187. No other economic conditions appear to be significantly associated with foundation giving in the same or the next quarter. However, GDP and PCE appear to positively significantly affect the number of foundation gifts 1 year later (see tables in Robustness section). This appears consistent with the countercyclical pattern we have noted.

To shed light on these results, we note that foundations may face different incentives and constraints in their million dollar giving decisions, compared with individuals. In particular, foundations may respond to adverse economic conditions by giving more million dollar gifts and grants. The increase in foundation giving during quarters when the United States is experiencing an economic recession seems to support this hypothesis. In addition, foundations may also respond countercyclically to macroeconomic factors if they are aware that individuals may decrease their giving during economic downturns; foundations may attempt to counteract this trend with their own giving.

Giving by corporations

Corporate million dollar giving, interestingly, showed no significant association with any of the macroeconomic indicators for the same quarter. Specifically, we did not find a significant relationship between corporate donations of one million dollars or more and any of the five major economic indicators tested. This indicates that, unlike other types of donors, corporate donors appear less responsive to macroeconomic conditions.

This finding is consistent with other studies that have noted that overall macroeconomic factors are less influential determinants of corporate giving compared with firm and industry variables (Amato & Amato, 2007; Urriolagoitia & Vernis, 2012). While individual donors appear responsive to macroeconomic conditions and foundation donors seem to have a countercyclical response, corporate donors simply appear less responsive than other donor types. Given this finding, our work does not provide additional support for Levy and Shatto’s “good citizen” hypothesis that the level of corporate contributions moves countercyclically and that these contributions actually rise when economic activity declines during a recession (Levy & Shatto, 1978).

Robustness checks on giving by donor type

For robustness, we examine the log amount donated (in dollars) within the quarter as the dependent variable. Looking at the log amount of all gifts donated, we see no relationship between our variables of interest and all gifts, nor any association when looking at gifts by source. There is a significant association between the S&P 500 index and the log amount donated (in dollars) to the arts and to health causes.

We also examine the log amount donated (in dollars) at the 25 percentile and 75 percentile levels as the key dependent variables to take into account the nonlinear distribution of million dollar and above gifts. Taken together, when we examine the log amount of million dollar giving (in dollars), we find that the amount donated is significantly associated with the S&P 500 index at the 25 percentile and 75 percentile levels of giving.

Economic Effects on Giving by Subsector

Our second research question asked to what extent million dollar giving to specific causes or subsectors changes in response to macroeconomic conditions. We are particularly interested in whether individuals, foundations, and corporations give more to support basic needs and human services during recessions.

Table 5 presents the baseline results of this analysis. The findings in Table 5 suggest that there is considerable variation in giving trends by subsector and donor type, and that the economic effects on those trends depend on the institutions and causes that receive million dollar gifts.

Table 5.

Impact of Economic Indicators on Incidence of Million-Dollar-Plus Giving to Causes by Subsector in the Same Quarter.

Dependent variable: Number of million-dollar-plus gifts per quarter by each donor type to each subsector
	All donors	Individuals	Corporations	Foundations
Panel A: Human services and public/societal benefit
β: S&P 500	−0.0410	0.0123	−0.0386	−0.0167
	(0.0384)	(0.0074)	(0.0318)	(0.0175)
	[−0.2290]	[0.3963]	[−0.2880]	[−0.2290]
β: GDP	−0.0176	0.0082	−0.0269	0.0009
	(0.0307)	(0.0075)	(0.0294)	(0.0208)
	[0.0459]	[0.9632]	[−0.7318]	[0.0459]
β: PCE	4.7677	0.9878	3.2193	0.9238
	(5.7430)	(1.2188)	(4.2840)	(3.2385)
	[0.3499]	[1.0599]	[0.7947]	[0.3499]
β: Unemployment	10.9096	−2.6705*	14.0851	−0.5308
	(10.8451)	(1.5676)	(10.0491)	(4.4027)
	[−0.0202]	[−0.8028]	[1.0084]	[−0.0202]
β: Recession	34.8395**	−3.6667	13.6422*	24.9136***
	(15.6552)	(4.5968)	(7.4885)	(7.6889)
	[0.7370]	[−0.2412]	[0.2102]	[0.7370]
Panel B: Arts, culture, humanities
β: S&P 500	0.0090	0.0180***	0.0005	−0.0078
	(0.0116)	(0.0066)	(0.0071)	(0.0119)
	[0.1465]	[0.8277]	[−0.0021]	[−0.2057]
β: GDP	0.0152	0.0137**	0.0132*	−0.0094
	(0.0119)	(0.0057)	(0.0069)	(0.0132)
	[0.9076]	[2.2925]	[2.6516]	[−0.9789]
β: PCE	0.4366	1.5820*	2.2293**	−2.8772
	(1.2726)	(0.9079)	(1.0477)	(1.2466)
	[0.1709]	[2.4213]	[4.1735]	[−2.5596]
β: Unemployment	−3.8131	−3.2401**	−3.8990**	2.8885
	(2.4405)	(1.4175)	(1.5824)	(2.8308)
	[−0.6075]	[−1.3918]	[−2.1103]	[0.7402]
β: Recession	3.7407	−1.4198	−1.1598	6.2099*
	(5.0894)	(2.4152)	(1.0191)	(3.6105)
	[0.1385]	[−0.1295]	[−0.0960]	[0.3028]
Panel C: Health
β: S&P 500	0.0052	0.0033	−0.0024	0.0040
	(0.0091)	(0.0049)	(0.0029)	(0.0062)
	[0.0656]	[0.1058]	[−0.1758]	[0.0949]
β: GDP	0.0074	0.0035	−0.0002	0.0035
	(0.0081)	(0.0054)	(0.0025)	(0.0052)
	[0.2719]	[0.3896]	[−0.0024]	[0.0716]
β: PCE	−0.6044	0.0092	−0.1412	−0.5559
	(1.3328)	(0.6379)	(0.3617)	(1.0040)
	[−0.5234]	[0.0433]	[−0.3669]	[−1.1034]
β: Unemployment	−0.1365	0.1216	1.0917	−0.9558
	(2.4463)	(1.7440)	(0.8648)	(1.3578)
	[0.0460]	[0.0146]	[0.7654]	[−0.1536]
β: Recession	4.6049	2.9630	2.1290	−0.5062
	(6.2417)	(2.9091)	(2.0370)	(2.7918)
	[0.1569]	[0.1817]	[0.3847]	[−0.0532]
Panel D: Higher education
β: S&P 500	0.0849	0.1117***	0.0048	−0.0315
	(0.0453)*	(0.0309)	(0.0111)	(0.0256)
	[0.3243]	[0.6177]	[0.0991]	[−0.3259]
β: GDP	0.1340**	0.1276**	−0.0001	0.0065
	(0.0612)	(0.0399)	(0.0100)	(0.0308)
	[1.8889]	[2.5594]	[−0.0701]	[0.2530]
β: PCE	7.8605	12.4368**	−0.6167	−3.9596
	(6.3767)	(4.7300)	(1.5553)	(3.1791)
	[1.0191]	[2.3230]	[−0.7291]	[−1.3610]
β: Unemployment	−37.1820**	−31.2054***	−0.5243	−5.4524
	(16.4692)	(10.4050)	(2.4208)	(8.9873)
	[−1.3455]	[−1.6036]	[−0.1228]	[−0.5249]
β: Recession	−4.0864	−1.3086	−3.5802	0.8025
	(22.6204)	(16.0732)	(6.0822)	(5.4817)
	[−0.0489]	[−0.0241]	[−0.2297]	[−0.0023]

Note. Each panel shows a separate OLS regression with a different independent variable as an indicator economic health (β), reporting the variable’s coefficient from a regression: Number of Gifts_ds,t = α + βIndicator_t + γYear_t + ζQuarter_t + ε_ds,t. Robust standard errors in parentheses. Standardized beta coefficients in brackets. S&P = Standard & Poor’s; GDP = gross domestic product; PCE = Personal Consumption Expenditures; OLS = ordinary least squares.

Significant at the 10% level. **Significant at the 5% level. ***Significant at the 1% level.

We analyze the economic effects on each subsector and discover that the results for human services, arts, health, and higher education organizations are striking. We focus on giving to human services organizations, which often provide basic needs assistance, and public and societal benefit services, as these include many institutions that provide housing, food, and other needs especially in demand during times of economic downturn. Such organizations include local United Way organizations, the American Red Cross, Salvation Army, and more. We find that giving by foundations and corporations to this area is countercyclical, as these organizations give more million-dollar-plus gifts to support human services and public/societal benefit services during recessionary periods. In quarters where there is at least 1 month of recession in the United States, corporate gifts to this area at the million-dollar-plus level rise by almost 14 gifts in the same quarter. This effect is even more evident in foundation giving: During quarters including at least 1 month of economic recession, foundation gifts to human services and public/societal benefit services rise by approximately 25 gifts in the same quarter. In the same quarter, an increase in the unemployment rate leads to a decrease in the number of individual gifts, and the association is significant for all subsectors except health. The results suggest that foundations are more responsive to the increasing demand for gifts to human services and public benefit organizations during periods of economic downturn, whereas individual giving to this area tends to decline when unemployment rises.

Compared with other subsectors, million dollar giving to arts and culture appears to show the most volatility. Relationships between the economic variables and the number of gifts directed toward arts, culture, and humanities organizations also appear to be significant for all donor types in the same quarter. There is an increase in foundation giving to these organizations during periods of recession. Corporations and individuals appear to give less to these organizations when the unemployment rate increases. Individual giving to the arts increases with positive changes in the S&P 500 index, GDP, and PCE index and decreases when unemployment rises. It may be that giving to these particular causes is considered a lower priority during periods of economic difficulty, particularly when compared with causes such as human services and health, which vary much less even when holding economic changes constant. We also consider some of the challenges in relying on data from public announcements on levels of giving, which may affect our interpretation about giving to specific subsectors. During times of recession, media reports may tend to focus relatively more attention giving to basic needs or the human service subsector than on giving to the arts. We consider the impact of the business cycle on media reporting patterns and consider comparing our results from media announcements with the annual tax data. The consistency of results based on tax records (although these are available on an annual basis) compared with the public announcement data increases our confidence in the validity of the results.

When analyzing the health subsector, we find no significant association between individual or foundation gifts at the million-dollar-plus level and any of the economic indicators analyzed.

Finally, economic indicators appear to be significantly associated with giving to the higher education subsector for individuals, but not for corporations or foundations. Giving from individuals to higher education acts procyclically and increases with growth in the S&P 500, the GDP, and the PCE, during the same quarter.

Robustness Checks

Lagged Dependent Variables

To account for the lagged effect of economic factors on million dollar giving, we conducted all analyses with the dependent variable lagged one quarter, as well as four quarters. These results are available upon request. The models examining giving by key donor types (individuals, foundations, and corporations) largely support the results in the nonlagged models.

When lagged 1 year, the relationship between giving by all donor types to the human services subsector appears to be affected by economic indicators. The human services subsector is the only one where, accounting for the 1 year lagged effect, three economic indicators are still significant (GDP, PCE index, and unemployment rate).

In the quarter following a recession, giving increases significantly, from all donor types combined, to public benefit and human services organizations. To reinforce this finding, giving by all donors combined, to arts, culture, and humanities organizations, decreases significantly in the quarter following an uptick in unemployment.

Poisson Model

One concern in our analysis is that the dependent variables are count variables and may not be normally distributed. A linear regression model may therefore not fully capture the underlying distribution of these types of data. To check the robustness of our results, we conducted the same analyses using Poisson regression rather than OLS. Poisson results for the donor type are shown in Table 6. We also conducted Poisson analyses for the subsectors, and for both donor type and subsector with the dependent variable lagged both one and four quarters, and these results are available upon request.

Table 6.

Impact of Economic Indicators on Incidence of Million-Dollar-Plus Giving in the Same Quarter Poisson Regression.

Dependent variable: Number of million-dollar-plus gifts per quarter by each donor type
	All donors	Individuals	Corporations	Foundations
Panel A: S&P 500
β: S&P 500	0.0607*	0.2052***	−0.0103	−0.0090
β: S&P 500	(0.0337)	(0.0406)	(0.097)	(0.0437)
Panel B: GDP
β: GDP	0.2388**	0.3298*	0.9146***	0.1491
β: GDP	(0.0977)	(0.1796)	(0.337)	(0.1235)
Panel C: PCE
β: PCE	0.1789**	0.2245	0.7759***	0.1083
β: PCE	(0.0812)	(0.1527)	(0.280)	(0.1065)
Panel D: Unemployment rate
β: Unemployment rate	−0.0370***	−0.0786***	0.0620	−0.0478*
β: Unemployment rate	(0.0140)	(0.0253)	(0.055)	(0.0248)
Panel E: Recession
β: Recession	0.0469***	0.0258	0.0934**	0.0685***
β: Recession	(0.0111)	(0.0248)	(0.039)	(0.0129)
Number of donations	19,077	7,092	3,003	8,982
N (Collapsed to quarters)	48	48	48	48

Note. The dependent variable and all independent variables are in log form, except the dummy variable for recession. Robust standard errors in parentheses. S&P = Standard & Poor’s; GDP = gross domestic product; PCE = Personal Consumption Expenditures.

Significant at the 10% level. **Significant at the 5% level. ***Significant at the 1% level.

Using Poisson regression to test results for the donor type (baseline model in Table 4), we find results are generally consistent with the baseline OLS model. In fact, the results from the Poisson regressions not only supported the previous results but also showed that the variables seem to have a more lasting impact further into the future. For individuals, both the stock market and unemployment rate appear to affect individual giving. Unlike our OLS model, the Poisson regression results for individuals showed that the PCE index becomes significant with a one-quarter lag and the recession becomes highly significant with a 1-year lag supporting the procyclical view of how the economy affects million dollar giving by individuals. For foundation giving, the Poisson regression results indicate that the unemployment rate lagged a quarter and a year, recession lagged a quarter, and S&P 500 lagged a year, and become significant with all supporting previous results showing foundation giving as countercyclical.

The Poisson model yielded more significant results for corporations. GDP and PCE are significantly positively associated with the incidence of million-dollar-plus corporate giving for all time periods. The results are mixed as the recession variable indicates countercyclical behavior and the GDP and PCE variables indicate procyclical behavior.

Using Poisson regression to test results for the subsector type (baseline model in Table 5) confirms many of our above findings, including that giving to human services and public benefit organizations is procyclical for individual donors and countercyclical for corporate and foundation donors. Unlike the OLS analysis, the Poisson regression results show that the key variables that influence gifts directed toward organizations in the arts, culture, and humanities by corporations and individuals are significant in each time period. Poisson results also indicate that foundation giving is positively affected by a recession in both the same quarter and the previous quarter. Overall, the Poisson results confirm that foundations increase gifts to arts, culture, and humanities during times of recession, but giving from corporations and individuals increases as GDP and PCE increase.

Overall, the Poisson results appear to support findings from our OLS results. We previously showed that economic indicators affect million dollar giving by individuals for the arts and higher education subsectors. Our Poisson results demonstrate that these impacts are felt across all four of these major subsectors. For foundation giving, the new results confirm a countercyclical pattern for human services, arts, and higher education subsectors and demonstrate that foundations seem most responsive to the recession indicator. These results further show that the impact of economic conditions on corporate giving is felt in both the short and long terms, and that PCE and GDP seem to elicit the most response in the number of gifts.

Number of Donors and the Amount Donated as Dependent Variables

One concern with the results on the impact of economic conditions is that we do not account for the influence of donors who give multiple gifts, as results are based on the number of gifts as the key dependent variable. To further check for the validity of our results, we examine the impact of economic conditions on the number of donors to avoid the influence of donors who give multiple gifts, as well as the impact of economic indicators and the number of donors in each donor category; results are shown in Table 7. We note that 90.9% of all single donors on the MDL gave only one gift. This approach allows us to determine whether a small fraction of donors are driving results by donating a large number of gifts during specific periods in our earlier model.

Table 7.

Impact of Economic Indicators on Number of Donors to Subsectors by All Donor Types.

Dependent variable: Number of donors giving gifts of 1 million or more per quarterDependent variable: Number of million-dollar-plus donors per quarter
	S&P 500	GDP	PCE	Unemployment	Recession
Arts/culture	0.017**	0.014**	1.580*	−3.431**	−1.235
	(0.006)	(0.005)	(0.895)	(1.350)	(2.352)
	[0.383]	[1.945]	[1.867]	[−1.232]	[−0.006]
Educ	0.010**	0.012**	1.550**	−2.436*	−1.321
	(0.004)	(0.005)	(0.758)	(1.226)	(1.337)
	[−0.028]	[0.488]	[0.523]	[−0.465]	[0.290]
Health	0.003	0.003	−0.031	0.273	2.630
	(0.005)	(0.005)	(0.638)	(1.758)	(2.971)
	[0.127]	[0.625]	[−0.280]	[−0.757]	[0.117]
Higher Ed	0.113***	0.129***	12.870***	−30.942***	−2.370
	(0.030)	(0.039)	(4.621)	(10.095)	(16.103)
	[0.453]	[2.392]	[1.916]	[−1.595]	[−0.032]
Int’l	0.000	0.000	0.803	−0.498	0.250
	(0.002)	(0.007)	(0.498)	(0.696)	(0.815)
	[−0.213]	[−0.051]	[−0.142]	[0.277]	[0.193]
Overseas	0.000	−0.004	−0.365	0.305	0.128
	(0.002)	(0.003)	(0.428)	(0.544)	(0.354)
	[0.116]	[0.407]	[0.015]	[0.531]	[−0.025]
Human services/Public society benefit	0.012	0.008	0.935	−2.657*	−4.383
	(0.007)	(0.007)	(1.192)	(1.499)	(4.655)
	[−0.089]	[0.020]	[1.217]	[0.574]	[0.228]
Int’l/overseas	−0.003	−0.006	−0.156	−0.362	0.603
	(0.002)	(0.005)	(0.696)	(0.781)	(0.686)
	[−0.009]	[0.186]	[0.006]	[0.486]	[0.065]

Note. Robust standard errors in parentheses.

Significant at the 10% level. **Significant at the 5% level. ***Significant at the 1% level.

Under these conditions, the impact of economic indicators on the number of donors appears very similar to the impact observed on the number of gifts.

We further analyze the way in which the specified economic indicators affect the number of donors to each subsector.

Interestingly, under this model, arts, culture, and humanities; education; and health seem to be the most responsive to economic conditions. The number of donors who give at the million dollar level to international and overseas organizations is not affected by domestic economic conditions. Finally, we also examine the log amount donated (in dollars) at the 25 percentile and 75 percentile levels as the key dependent variable to take into account the skewed distribution of million dollar and above gifts. Taken together, when we examine the log amount of million dollar giving (in dollars), we find that the amount donated is significantly associated with the S&P 500 index at the 25 percentile and 75 percentile levels of giving.

Conclusion

There is growing public and policy interest in the potential for philanthropy to meet societal challenges. Despite a growing number of empirical studies on charitable giving, there have been very few studies that examine million dollar gifts, due in part to data limitations. (There is a small but growing body of research and data on million dollar giving, including The Coutts Million Pound Donors Report (Breeze, 2009) and Coutts Million Dollar Donors Reports (years 2010-2015) (Indiana University Lilly Family School of Philanthropy)). Using a newly available data set on million dollar giving, we find important differences in how million-dollar-plus giving is influenced by economic conditions among the three primary types of donors. Individual donors are particularly responsive to underlying economic conditions as measured by the S&P 500 index, as well as other key economic variables. Similarly, individual giving at this level is negatively correlated with unemployment and positively correlated with GDP growth.

In contrast to our findings on individual giving, we find that million dollar giving by foundations tends to be countercyclical. During quarters in which there is at least 1 month of recession, foundations appear to increase their giving of million-dollar-plus gifts and grants. An important finding is that the incidence of corporate million dollar giving is not significantly associated with overall macroeconomic factors.

Relevant to public policy debates, we find that economic conditions influence giving by subsector in significant ways. In particular, increased gifts are targeted toward human service and public benefit organizations. Economic indicators such as the S&P 500 index, GDP, and PCE also indicate more significant effects on donations to these organizations. These results suggest that the foundation sector may seek to address adverse economic conditions and societal and human needs through their grantmaking.

In the aftermath of the Great Recession, these findings are valuable to researchers, nonprofit leaders, policy makers and practitioners because understanding the overall relationship between giving and the economy will inform the development of effective fundraising or donor engagement strategies. During adverse economic shocks, it is important for nonprofits to know how to allocate their resources to various segments of donors, recognizing that individuals, foundations, and corporations may respond differently in the face of changes in economic conditions.

Footnotes

Authors’ Note

Thanks to Jon Bergdoll, Jason Ward, Michael Copple, John DeWolf, and Elizabeth Farris for their contributions to the research.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research was completed with funding from the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect official positions or policies of the Bill & Melinda Gates Foundation.

Author Biographies

Una O. Osili, PhD, is professor of economics and associate dean for Research and International Programs at the Indiana University Lilly Family School of Philanthropy.

Jacqueline Ackerman, MPA, is assistant director for Research and Partnerships for the Women’s Philanthropy Institute at the Indiana University Lilly Family School of Philanthropy.

Yannan Li is a PhD candidate in Philanthropic Studies and a former research assistant at the Indiana University Lilly Family School of Philanthropy.

References

Amato

L. H.

Amato

C. H.

(2007). The effects of firm size and industry on corporate giving. Journal of Business Ethics, 72, 229-241.

Andreoni

(1993). An experimental test of the public-goods crowding-out hypothesis. American Economic Review, 83, 1317-1327.

Andreoni

Petrie

(2004). Public goods experiments without confidentiality: A glimpse into fund-raising. Journal of Public Economics, 88, 1605-1623.

Breeze

(2009). The Coutts million pound donors report. Canterbury, UK: The University of Kent.

Congressional Budget Office. (2011). Options for changing the tax treatment of charitable deductions. Washington, DC: Congress of the United States.

Coutts. (2016). 2016 million dollar donors report. Author. Retrieved from http://philanthropy.coutts.com/en/reports/2016/united-states/findings.html

Davies

J. B.

(1981). Uncertain lifetime, consumption, and dissaving in retirement. Journal of Political Economy, 89, 561-577.

DellaVigna

List

J. A.

Malmendier

(2009). Testing for altruism and social pressure in charitable giving (NBER Working Paper No. 15629). Stanford, CA: National Bureau of Economic Research.

Drezner

N. D.

(2006). Recessions and tax-cuts: Economic cycles’ impact on individual giving, philanthropy, and higher education. International Journal of Educational Advancement, 4, 289-305.

10.

Drezner

N. D.

(2010). Fundraising in a time of economic downturn: Theory, practice and implications—An editorial call to action. International Journal of Education Advancement, 9, 191-195.

11.

Duncan

(2004). A theory of impact philanthropy. Journal of Public Economics, 88, 2159-2180.

12.

Foundation Center. (2010). Moving beyond the economic crisis (Research Advisory). New York, NY: Foundation Center.

13.

Galaskiewicz

(1997). An urban grants economy revisited: Corporate charitable contributions in the twin cities, 1979-81, 1987-89. Administrative Science Quarterly, 42, 445-471.

14.

Giving USA Foundation. (2009). Giving recovery after economic depression or recession. Giving USA Spotlight, Issue 2, 1-9.

15.

Giving USA: The Annual Report on Philanthropy for the Year 2013. (2014). Chicago, IL: Giving USA Foundation.

16.

Giving USA: The Annual Report on Philanthropy for the Year 2015. (2016). Chicago, IL: Giving USA Foundation.

17.

Giving USA: The Annual Report on Philanthropy for the Year 2016. (2017). Chicago, IL: Giving USA Foundation.

18.

Havens

O’Herlihy

Schervish

(2006). Charitable giving: How much, by whom, to what, and how? In Powell

Steinberg

(Eds.), The nonprofit sector: A research handbook (pp. 542-567). New Haven, CT: Yale University Press.

19.

Hurd

M. D.

(1989). Mortality risk and bequests. Econometrica, 57, 779-813.

20.

Indiana University Lilly Family School of Philanthropy. (2012). Million dollar list. Available from http://www.milliondollarlist.org

21.

Internal Revenue Service. (2016). SOI tax stats–Statistics of income. Retrieved from https://www.irs.gov/uac/soi-tax-stats-statistics-of-income

22.

Joulfaian

(2005). Choosing between gifts and bequests: How taxes affect the timing of wealth transfers. Journal of Public Economics, 89, 2069-2091.

23.

Katz

S. N.

(2005). What does it mean to say that philanthropy is “effective”? The philanthropists’ new clothes. Proceedings of the American Philosophical Society, 149, 123-131.

24.

Kingma

B. R.

(1989). An accurate measurement of the crowd-out effect, income effect, and price effect for charitable contributions. Journal of Political Economy, 97, 1197-1207.

25.

Kitzmueller

Shimshack

(2012). Economic perspectives on corporate social responsibility. Journal of Economic Literature, 50, 51-84.

26.

Levy

F. K.

Shatto

G. M.

(1978). The evaluation of corporate contributions. Public Choice, 33(1), 19-28.

27.

Lew

Wójcik

(2009). The cultures of philanthropy: Private foundation governance in the USA, the UK, Germany, and Japan. doi:10.2139/ssrn.1508858

28.

List

J. A.

Peysakhovich

(2011). Charitable donations are more responsive to stock market booms than busts. Economics Letters, 110, 166-169.

29.

Long

S. H.

(1976). Social pressure and contributions to health charities. Public Choice, 28, 55-66.

30.

Moir

Taffler

R. J.

(2004). Does corporate philanthropy exist? Business giving to the arts in the U.K. Journal of Business Ethics, 54, 149-161.

31.

National Bureau of Economic Research. (2012). US business cycle expansions and contractions. Retrieved from http://www.nber.org/cycles.html

32.

Olsen-Phillips

(2016, October 27). How the 2016 Philanthropy 400 was compiled. The Chronicle of Philanthropy. Retrieved from https://www.philanthropy.com/article/How-the-2016-Philanthropy-400/238166

33.

Piketty

Saez

(2006). The evolution of top incomes: A historical and international perspective (No. w11955). Stanford, CA: National Bureau of Economic Research.

34.

Reich

Wimer

Mohamed

Jambulapati

(2011). Has the great recession made Americans stingier? In Grusky

D. B.

Western

Wimer

(Eds.), The great recession (pp. 294-313). New York, NY: Russell Sage Foundation.

35.

Rooney

P. M.

Steinberg

K. S.

Schervish

P. G.

(2001). A methodological comparison of giving surveys: Indiana as a test case. Nonprofit and Voluntary Sector Quarterly, 30, 551-568.

36.

Samek

A. S.

Sheremeta

R. M.

(2014). Recognizing contributors: An experiment on public goods. Experimental Economics, 17, 673-690.

37.

Schervish

P. G.

Havens

J. J.

(2001). Wealth and the commonwealth: New findings on wherewithal and philanthropy. Nonprofit and Voluntary Sector Quarterly, 30, 5-25.

38.

Seifert

Morris

S. A.

Bartkus

B. R.

(2003). Comparing big givers and small givers: Financial correlates of corporate philanthropy. Journal of Business Ethics, 45, 195-211.

39.

Urriolagoitia

Vernis

(2012). May the economic downturn affect corporate philanthropy? Exploring the contribution trends in Spanish and U.S. companies. Nonprofit and Voluntary Sector Quarterly, 41, 759-785.

40.

U.S. Bureau of Economic Analysis. (2012). GDP and the National Income and Product Account (NIPA) historical tables. Retrieved from https://www.bea.gov/system/files/2018-08/SNTables_0.pdf

41.

U.S. Bureau of Labor Statistics. (2012). Local force statistics from the current population survey. Retrieved from http://www.bls.gov/cps/

42.

Wilhelm

M. O.

(2006). New data on charitable giving in the PSID. Economics Letters, 92, 26-31.

43.

Wilhelm

M. O.

(2007). The quality and comparability of survey data on charitable giving. Nonprofit and Voluntary Sector Quarterly, 36, 65-84.

44.

Yahoo! Finance. (2012). S&P 500 historical prices. Retrieved from http://finance.yahoo.com/q/hp?s=^GSPC+Historical+Prices