Saving with a Social Impact: Evidence from Trento Province

In this paper, we investigate the determinants of investing, focusing on its potential social impact. In particular, we consider whether there is room for expanding impact investing through social savings. The increasing trend in the demand for social finance makes the topic of unique interest, particularly when data on preferences for social saving can be collected at the individual level. We investigate the determinants and drivers of saving with a social goal running a survey conducted in Trentino-Alto Adige in which respondents are asked to allocate their portfolio to possible social investments. In line with the evidence collected in the Netherlands by Riedl and Smeets (2014), our results show a strong preference for a lower return, with the condition that the return is invested in a community programme, and little interest in the monetary return of the investment. Respondents are either inclined to put their entire portfolio into saving for the community or not to invest at all. This result suggests that there is a consistent demand, only partially accommodated by the supply, for financial products investing in the community, rather than for a monetary return.


Introduction
This study deals with an expanding phenomenon encompassing social financing: saving with a social impact. Socially responsible investment has an ethical component over and above financial gain, being aimed at enterprises with best practices that includes sustainability. To this end, the investment can be based on screening strategies, which may be positive, selecting investments that are coherent with the principles of responsibility, or negative, excluding investments in enterprises that operate in non-ethical sectors, such as tobacco and gambling.
In the world of finance, the will to maintain investment principles is intertwined with the duty to consider the social side. The classic risk-versus-gain trade-off is accompanied by the social sustainability of investments. Socially responsible funds select investments according to ESG (environmental, social and governance) principles, a trend that is encouraged by the UN 2030 Agenda [1]. Beal et al. [2] argue that socially responsible investment cannot be made with the sole purpose of seeking extra-financial performance, as the financial gain from this type of investment is anything but proven. Riedl and Smeets [3] show that many investors in socially responsible assets expect lower returns for the same level of risk. The reason why they still choose to have investments controlled by other investments has to do with their social preferences. The social identity of investors

Literature Framework
Social finance is the branch of finance in which the management of money aims to obtain not only a financial but also a social return. Battaglia et al. [7] identify it as the set of alternative approaches to investment and financing decisions intending to generate a positive impact on society and/or on the environment, thus also including green finance in social finance. This characterization of social finance makes it an example of ethical finance, the origins of which, rooted in the past, are sometimes intertwined with religious principles [8].
Within social finance, socially responsible investments (SRI) have been attracting the interest of many economic agents in that they represent a useful financing tool from the viewpoint of firms and an interesting investment opportunity for investors, including retail ones. Accordingly, the academic literature adopts many different perspectives. To frame our contribution, we adopt the personal finance perspective, from which the academic literature addresses a few related questions.
First, what are the drivers of social investors? An answer can be found by setting the issue in a theoretical framework in which the individual's utility function depends on both wealth and non-wealth returns, the latter capturing the socially responsible dimensions of the decision. Among others, Bollen [9] tests the differences in behaviour between investors in socially responsible (SR) mutual funds and investors in conventional funds; the results are consistent with a multi-attribute utility function, with investors not only looking at the risk-return trade-off but also gaining direct utility from the socially responsible attributes of the funds-the so-called intrinsic motivation, that is, the value of giving per se [10]. In the same spirit, Beal et al. [2] provide three non-exhaustive and non-exclusive motivations for ethical investment: superior financial returns (consistent with traditional finance Sustainability 2020, 12, 8363 3 of 15 theory), non-wealth returns and social change. Based on the theoretical model of utility of wealth and morality by Levitt and List [11] and using a natural field experiment, Døskeland and Pedersen [12] show that wealth framing is more effective than moral framing in inducing investors to engage in SRI. An experimental approach is also taken by Pasewark and Riley [13], who ask individuals to choose between bonds issued by a tobacco company and bonds issued by a firm outside the tobacco industry and conclude that the personal values of the investor affect the investment decisions.
A second issue addressed in the literature is the historical performance of SRI compared with conventional investments and hence the potential existence of an "ethical penalty". In this connection, the literature is not univocal, given that a few studies find that SRI offers lower risk-adjusted returns and others obtain the opposite conclusion. For instance, Renneboog et al. [14] find that SRI funds in European, North American and Asia-Pacific countries underperform with respect to conventional ones, implying that SRI investors pay a price for their socially responsible choice. On the other hand, Bauer et al. [15], using a database of German, UK and US ethical mutual funds, do not find significant differences in risk-adjusted returns between ethical and conventional funds. By contrast, Martí-Ballester [16] analyses the biotechnology and healthcare sectors, finding out that SRI investments outperform traditional mutual funds. However, recent evidence clearly underlines that their performance usually depends on the market state; indeed, Dopierała et al. [17] evaluate the performance of Scandinavian climate-themed funds, assessing the importance of geographical and industrial asset allocations.
Given the increasing interest in green bonds, recent studies focus on the determinants of the higher premium of green bonds (e.g., Zerbib [5]; and, with a focus on China, Wang et al. [18]).
A related question is the willingness of investors to accept a lower return, should it be the case: Renneboog et al.'s [14] review stresses that existing studies hint at but do not univocally prove the willingness of agents to accept a lower return in exchange for social or ethical goals. Riedl and Smeets [3] highlight social preferences as the main driver of investment in SRI despite the expectation of a lower return. Rossi et al. [19] show that social investors are willing to pay a price to be socially responsible rather than needing a little nudge, such as a gift; however, particularly at the intensive margin, the stated demand for SRI funds is sensitive to the return penalty.
The third question in the literature concerns SRI investors' profile. Bauer and Smeets [20] and Hoque et al. [21] provide evidence supporting high levels of social identification among young, highly educated and low-wealth investors. The roles of gender and education are also highlighted by Nilsson [22], who further shows that social investors are driven not only by altruistic motives but also by the idea that ethical mutual funds have an average or better than average performance. Hood et al.'s [23] recent study examined heterogeneities among socially conscious investors, emphasizing the different preferences for social investments across gender, age, religion and groups with different political affiliations.
Therefore, on the basis of a collaboration protocol between the University of Trento, the Employment Agency of the Autonomous Province of Trento and ISPAT (the Statistical Office of the Autonomous Province of Trento), we designed a survey to examine not only the financial behaviour of savers but also the relationship between their behaviours, their local community and social issues. To the best of our knowledge, no other study in the literature has taken this threefold perspective.

The Framework of Socially Responsible Finance
Socially responsible investments accounted for 2% of the total investments in Europe and amounted to €158 billion in 2016, continuously increasing to a peak of 16% between 2015 and 2016 ( Figure 1). Unlike the market trend, the sector expanded throughout the economic downturn and immediately after it. Sustainable financing is here to stay and is relatively substantial, amounting to a third of the global GDP (Vigeo Eiris, http://vigeo-eiris.com/wp-content/uploads/2016/11/161020-Green-Social-and-Ethical-funds-in-Europe-2016_Final-Compatibilit...-1.pdf, 2016).

The Framework of Socially Responsible Finance
Socially responsible investments accounted for 2% of the total investments in Europe and amounted to €158 billion in 2016, continuously increasing to a peak of 16% between 2015 and 2016 ( Figure 1). Unlike the market trend, the sector expanded throughout the economic downturn and immediately after it. Sustainable financing is here to stay and is relatively substantial, amounting to a third of the global GDP (Vigeo Eiris, http://vigeo-eiris.com/wp-content/uploads/2016/11/161020-Green-Social-and-Ethical-funds-in-Europe-2016_Final-Compatibilit...-1.pdf, 2016). The Global Sustainable Investment Review Error! Reference source not found. (http://www.gsialliance.org/wp-content/uploads/2019/03/GSIR_Review2018.3.28.pdf) specifies the criteria applicable to sustainable investment according to environmental, social and governance principles: (1) exclusion of securities from the investible universe, (2) regulatory screening, (3) active stocks, (4) ESG integration, (5) selection of "best-in-class" securities, (6) sustainable investments (sustainability themed) and (7) impact investing (positive impact investments). These criteria include ethical and other considerations, with an investment strategy that is monitored for non-responsible behaviour and involves the selection of investments biased towards sustainability, the latter having an intrinsic intertemporal dimension that is not limited to the short term. However, exclusion remains the predominant factor in sustainable investments, with tobacco and armaments being avoided by fund managers when deciding on investment options.
According to the Report of Consob Error! Reference source not found. (http://www.consob.it/documents/46180/46181/rf2019.pdf/b3a1763b-a869-4aca-8930-9ae370a0aa90) on the investment choices of Italians, SRI are still little known by Italian investors, who reported in only 5% of cases that they have such products in their investment portfolios. This is probably due to a scarce knowledge of SRIs: 40% of the respondents admitted to having heard of them, but only 5% declared that they have good knowledge about them. Almost 40% of the sample was estimated to be potentially interested in SRIs, while the lack of interest is attributable to a dearth of savings for investment, not having received proposals on SRIs or mistrusting them. The potential and effective demand for SRIs seems to be accentuated among the wealthier subjects with a higher level of education and financial knowledge, sensitivity towards social issues and greater attention to longterm performance.

Preferences stated for Social Issues
The Italians define themselves as generous towards other people. In fact, the well-being of others and helping fellow citizens are considered important by the vast majority, as evidenced by the 2012 European Social Survey (ESS, 2012), where the respondents had to state whether they disagree or The Global Sustainable Investment Review [24] (http://www.gsi-alliance.org/wp-content/uploads/ 2019/03/GSIR_Review2018.3.28.pdf) specifies the criteria applicable to sustainable investment according to environmental, social and governance principles: (1) exclusion of securities from the investible universe, (2) regulatory screening, (3) active stocks, (4) ESG integration, (5) selection of "best-in-class" securities, (6) sustainable investments (sustainability themed) and (7) impact investing (positive impact investments). These criteria include ethical and other considerations, with an investment strategy that is monitored for non-responsible behaviour and involves the selection of investments biased towards sustainability, the latter having an intrinsic intertemporal dimension that is not limited to the short term. However, exclusion remains the predominant factor in sustainable investments, with tobacco and armaments being avoided by fund managers when deciding on investment options.
According to the Report of Consob [25] (http://www.consob.it/documents/46180/46181/rf2019.pdf/ b3a1763b-a869-4aca-8930-9ae370a0aa90) on the investment choices of Italians, SRI are still little known by Italian investors, who reported in only 5% of cases that they have such products in their investment portfolios. This is probably due to a scarce knowledge of SRIs: 40% of the respondents admitted to having heard of them, but only 5% declared that they have good knowledge about them. Almost 40% of the sample was estimated to be potentially interested in SRIs, while the lack of interest is attributable to a dearth of savings for investment, not having received proposals on SRIs or mistrusting them. The potential and effective demand for SRIs seems to be accentuated among the wealthier subjects with a higher level of education and financial knowledge, sensitivity towards social issues and greater attention to long-term performance.

Preferences stated for Social Issues
The Italians define themselves as generous towards other people. In fact, the well-being of others and helping fellow citizens are considered important by the vast majority, as evidenced by the 2012 European Social Survey (ESS, 2012), where the respondents had to state whether they disagree or agree with some statements. According to the survey, almost 80% identify with the idea that it is important to help others and take care of them (see Table 1). How many of these statements translate into concrete action? The survey includes information on voluntary activities, which can be interpreted as concrete action to care for others. Table 2 shows lower percentages, albeit always significant. The majority of the respondents have never performed voluntary work (56%), whereas about 30% have undertaken voluntary work for one semester. If we look at the gender dimension, participation in voluntary activities is more common among men (49% of men versus 39% of women), probably because men have more free time than women, who still spend a lot of time on housework (the highest number of hours in Europe, as found by Della Giusta [26]).
Time spent on voluntary work is matched with donations of money, both aiming to increase the welfare of others without personal gain. With donations, involvement in the financed activity is often long distance and the recipient is not known directly. With the same purpose, the dimension of time, through volunteering, joins the monetary dimension through monetary donations: both are aimed at increasing the well-being of others without any personal reward. For monetary donations, individual involvement in the subsidized activity is often remote, targeting realities that are not directly known.
Currently, there is an increasing trend to donate locally in terms of both the collection and the employment of funds. Crowdfunding is a significant example. The purpose of crowdfunding may be to finance or donate to cultural or social projects almost/entirely without returns. Unlike conventional donations, crowdfunding is aimed at well-defined projects that are clearly explained to potential investors. The sum needed is publicized together with the proposal, and the project in question is financed only if the stated amount is reached. Donor involvement is increased by the fact that the place and activity are specified in the description (for example, the refurbishment of an abandoned building). Remuneration is often symbolic, as in the case of a gadget, although proportional to the donation. The numbers of people participating in crowdfunding are steadily increasing.
A new sector with a promising future is that of socially responsible savings. Unlike crowdfunding, ethical savings are a conventional product that presupposes a part to be returned. They constitute a "suboptimal" product compared with those on the market, deliberately and transparently offering reduced gains for the purpose clearly stated in the specifications. When buying an ethical product, the investor is well aware that this implies reduced gains. An example of an ethical and socially responsible product is the acquisition of social bonds, which only recently appeared on the market in Italy but are rapidly gaining in popularity.
Social bonds have both the characteristics of a classic financial product and a transfer component linked to a project known ex ante by the underwriter. Social bonds therefore contain both the non-profit and the traditional dimensions. Like the non-profit sector, which has increased by 40% considering the associated activities and workers, social bonds have grown by almost twenty times in three years (ECF Frankfurt (Germany), 2018) (https://www.ecb.europa.eu/paym/groups/pdf/bmcg/180206/2018-02-06_-_ BMCG_-_Item_3_-_Update_on_the_green_and_social_bond_market_-_ICMA.pdf).
It is useful to remember that such explosive growth takes place in spite of the fact that the financial product performs at below-market yield, thus being sub-optimal as an investment. As shown in Figure 2, the yield can in fact be lower than that of conventional bonds by as much as 4 percentage points. If we look at the European framework, the difference in performance between conventional and green or social bonds has reached almost 4 percentage points (with the European Investment Bank as the issuing bank), while the minimum yield differential has been about 0.3 percentage points for the NRW Bank or 0.02 percentage points in the recent study by [5]. and socially responsible product is the acquisition of social bonds, which only recently appeared on the market in Italy but are rapidly gaining in popularity.
Social bonds have both the characteristics of a classic financial product and a transfer component linked to a project known ex ante by the underwriter. Social bonds therefore contain both the nonprofit and the traditional dimensions. Like the non-profit sector, which has increased by 40% considering the associated activities and workers, social bonds have grown by almost twenty times in three years (ECF Frankfurt (Germany), 2018) (https://www.ecb.europa.eu/paym/groups/pdf/bmcg/180206/2018-02-06_-_BMCG_-_Item_3_-_Update_on_the_green_and_social_bond_market_-_ICMA.pdf).
It is useful to remember that such explosive growth takes place in spite of the fact that the financial product performs at below-market yield, thus being sub-optimal as an investment. As shown in Figure 2, the yield can in fact be lower than that of conventional bonds by as much as 4 percentage points. If we look at the European framework, the difference in performance between conventional and green or social bonds has reached almost 4 percentage points (with the European Investment Bank as the issuing bank), while the minimum yield differential has been about 0.3 percentage points for the NRW Bank or 0.02 percentage points in the recent study by Error! Reference source not found..

Data
For the empirical analysis, we use an "omnibus" survey conducted by ISPAT, the Statistical Office of the Autonomous Province of Trento, specialized in promoting and coordinating the production and dissemination of statistics of interest to the provincial administration; it also carries out studies, censuses, surveys and statistical elaborations. On the basis of a collaboration protocol between the University of Trento, the Employment Agency of the Autonomous Province of Trento and ISPAT, we designed a special module with questions that enrich the data set with new information on social attitudes towards savings. Respondents are members of a survey conducted by ISPAT. Interviews are usually carried out by telephone. This method of interviewing is known as CATI (Computer Assisted Telephone Interviewing). The survey was delivered during spring 2019 and covers a representative sample of 1073 adult individuals of Trento, half of whom are female. A random sample of telephone numbers relating to families residing in Trentino was extracted. An average of three reserves were available for each name. A total number of 1061 valid interviews have been carried out, by randomly selecting the adult person to be interviewed within the family, using the "birthday" method, that is, choosing the adult whose birthday appears to be the most recent with respect to the day of the interview. A faithful copy of the questionnaire, originally administered in Italian, translated into English is reported in Appendix B.

Data
For the empirical analysis, we use an "omnibus" survey conducted by ISPAT, the Statistical Office of the Autonomous Province of Trento, specialized in promoting and coordinating the production and dissemination of statistics of interest to the provincial administration; it also carries out studies, censuses, surveys and statistical elaborations. On the basis of a collaboration protocol between the University of Trento, the Employment Agency of the Autonomous Province of Trento and ISPAT, we designed a special module with questions that enrich the data set with new information on social attitudes towards savings. Respondents are members of a survey conducted by ISPAT. Interviews are usually carried out by telephone. This method of interviewing is known as CATI (Computer Assisted Telephone Interviewing). The survey was delivered during spring 2019 and covers a representative sample of 1073 adult individuals of Trento, half of whom are female. A random sample of telephone numbers relating to families residing in Trentino was extracted. An average of three reserves were available for each name. A total number of 1061 valid interviews have been carried out, by randomly selecting the adult person to be interviewed within the family, using the "birthday" method, that is, choosing the adult whose birthday appears to be the most recent with respect to the day of the interview. A faithful copy of the questionnaire, originally administered in Italian, translated into English is reported in Appendix B.
The administered questionnaire focuses on ascertaining the orientation towards investments with a social component. Given that the Italian savings rate is still consistent, it is interesting to explore whether there is a latent component of savers who would turn towards investments with a high social content, making a visible impact in the context of the community to which they belong.
Half of the sample, around 53% of respondents, declared that they were not saving ( Table 3). The attitude towards saving differs substantially according to the respondents' age group, with individuals in the middle of their life cycle tending to accumulate the most, as the life cycle theory predicts. Among those who save, the highest percentage is found in individuals aged between 33 and 65 years old (343 respondents). Of the respondents, 13% declared that they knew about crowdfunding (141 individuals), a phenomenon that is on the rise but still unknown to most. Of the 141 individuals who knew about crowdfunding, only 35 donated money while 9 invested money. If we compare this with the numbers for donations in general, they are much more present in the habits of Trentino people. In fact, the percentage of those who made at least one donation stood at around 30% (316 respondents), against the 35 people who donated through crowdfunding (Table 4a,b). The crowdfunding phenomenon mainly concerns those who are of working age, without great differences in percentage terms between the youngest (under 33 years) and those in the range from 33 to 65 years old. Looking at the dimensions of gender, education and preferences (Table 5a,b) regarding socially responsible saving, we found that, while no significant differences are found in gender, 20% of the better-educated respondents have heard of, and know about, crowdfunding (the exact wording of the question was: "Have you ever heard of crowdfunding, in other words of raising funds through internet websites to support cultural, social or financing projects in the form of straight donations or yielding financial returns?") whereas only 8% of those with below secondary education diplomas are acquainted with crowdfunding. The education level is positively correlated with the income level, so the implicit link highlighted could be income related. Unfortunately, ISPAT did not provide us with any detailed information about income (due to a strict privacy policy), so education represents the only source of a proxy for income.

Interest in Social Savings in the Trento Area
Interesting results emerge from observing the percentage of individuals who declare themselves to be interested in an "ethical" product, that is, a type of investment that yields no returns for investors but the gains of which are invested in a project with a social background for the community (the exact wording referring to the financial product in question was as follows: "For Product B, with zero yield, the issuing institution will use the resulting interest, i.e., the €100, to finance a project for the benefit of the community, e.g., a grant to the local poor, a subsidy to the disabled, the extension of a hospital, etc."). In fact, 32.5% of the sample declared themselves to be in favour of such investments compared with investing in conventional products. The exact wording of the question was as follows: "Supposing that €10,000 of your financial assets (real estate excluded) are invested in a yearly savings account. There are two investment options: Product A, conventional, paying interest of 1% (€100 after one year). Product B, in which you will not receive any interest, but the issuing institution will use the "missed" interest, that is, the €100, to finance a project for the benefit of the community, (e.g., a grant to the local poor, a subsidy for the disabled, the extension of a hospital, etc.). How much of the €10,000 would you invest in these two products?" The possible answers were: (1) all in product A, (2) half in A and half in B and (3) all in B. We also offered the option of "not investing anything" as well as the option of not responding.
The question clearly highlighted the inferiority of the "generous" product. Nonetheless, the majority of would-be investors would opt for the social product, even at a loss, compared with the market returns (about 30%). The absolute majority of the sample denoted a lack of interest: about 30% would not invest at all, while 20% did not answer, indicating that maybe a segment of the sample was not even interested in associating philanthropy at the local community level with financial products.

Empirical Analysis and Results
To move beyond the detailed descriptive statistics provided in the previous section, we performed an empirical analysis to assess the determinants of some of the above aspects, such as the propensity to donate or the propensity to invest in ethical products (the main objective of this study).
We estimated the following equation: where the dependent variable is a set of dummy variables, explained below, while the vector of regressors X is a set of sociodemographic variables that were available in the survey. Given the binary nature of the variables, we estimated Equation (1)  All the dependent variables were constructed as dummy variables that assume the value 1 when the respondent answers the question in the survey positively and 0 otherwise. For the variable "Preference for investing in ethical products", the dummy takes the value 1 when the individual answers that he or she is willing to invest everything in Product B (see Table 6 for details). A detailed description of each variable can be found in the Table A1 in Appendix A. Table 6. Interest in ethical products.
Finally, supposing you invest €10,000 of your assets (real estate excluded) in a yearly savings/deposit account. Two investment options were considered: Product A, conventional, implying a 1% interest gain (€100 interest after one year) Product B, zero yield, but the issuing institution will use the resulting interest, i.e., the €100, to finance a project for the benefit of the community, e.g., a grant to the local poor, a subsidy to the disabled, the extension of a hospital...). How much of the €10,000 would you invest in these two products?

Percent values
All in product A, yielding €100 of interests 8.9 Half in A and half in B, receiving €50 in interests 8.4 All in B, yielding zero interest, the entire gain being devolved to a project benefiting the community 32.5 Would We used as the main explanatory variables demographic controls such as gender, education level and age. They were constructed as follows: • Female: a dummy variable that takes the value 1 for female individuals and 0 otherwise; • Higheduc: a dummy variable with the value 1 if the individual has an education level equal to "diploma, degree level or above"; • Eta_*: dummy variables based on the age of the respondents (expressed through categories); in detail, there are three classes: 18 to 32 years old, 33 to 65 years old and 66 and above (age_1, age_2 and age_3, respectively).
All the results are reported in Table 7 (each column corresponds to a different dependent variable; we also performed an OLS estimation, the results of which are available on request). Overall, the results appear to be consistent with the existing literature, pointing to a positive link between interest in SRI products and education level (Nilsson [22] and Consob Report [25]). Indeed, the results show that a higher level of education has a positive impact on the knowledge of crowdfunding; this could mean that the most educated individuals are likely to be better informed (first column of Table 7). Let us specify that crowdfunding has been included among the different possibilities of financial products with donations, as crowdfunding can consist of equity, peer-to-peer lending and donations. The latter is actually the biggest actor in crowdfunding. For this reason, we wanted to include crowdfunding, as it specifically associates the money invested and the goal that it aims to fund, which is the common feature with the social bond, in line with our previous paper focused on the Netherlands [19]. Education is the closest proxy to income in the data, which provide no information on income, either in interval or in absolute values. The positive impact of education is also significant in terms of donations through crowdfunding; moreover, the female gender has a negative impact (second column of Table 7). Probit estimation models. Marginal effects reported. Standard errors in parentheses. * p < 0.10; ** p < 0.05; *** p < 0.01.
As for having invested through crowdfunding or donated a sum of money, there is no significant evidence. However, it should be stressed that the number of answers and positive answers for these questions was quite small. Furthermore, as proposed in the work by Marchegiani [27], crowdfunding could represent a cultural and creative chance, because people, once exposed to the issues of social savings, do want to contribute independently of the monetary returns.
Analysing the results in the third column of Table 7, there is a negative impact linked to the gender of the individuals on their savings; therefore, women tend to save less than men. A possible explanation for this could also be related to work: women tend to work less than men, and this could have an impact on their savings. This result is in line with the existing literature that suggests that women tend to save less if, for example, there are diseconomies in the marriage [28]. As regards the propensity to save in the future (the fourth column of Table 7), it appears that the only relevant variable is the level of education, which can probably be interpreted in light of greater earning prospects.
Finally, the most interesting result is probably that in the last column of Table 7. The table shows the regression results, in which the dependent variable is a dummy that assumes the value 1 if the individual is deeply interested in investing everything in a product defined as "ethical" or a product that does not give a return but that contributes to a project for the community. Age is another crucial factor: it can in fact be noted that the categories of respondents aged between 33 and 65 years and 66 years and over are those that tend to invest less in ethical financial products: this could simply indicate that older people probably tend to prefer more traditional products than the "younger" categories. An explanation for this evidence could be an implicit medium-long-term characteristic of investments which, apart from financial characteristics, also include ethical and, more generally, sustainable products. The results in this case also prove to be coherent with the national findings indicated in the Consob Report [25]: among the individual traits of SRI investors, sensitivity towards social issues is not the only noticeable aspect; a greater orientation towards long-term performance is also apparent.
The analysis would certainly benefit from income-related variables; however, we are aware that data on income and wealth are usually missing in the responses. It would also be extremely important to clarify which kinds of projects make the investors more interested in a social bond, for a better tailoring of the financial product.

Conclusions
We reviewed socially responsible financing from the viewpoint of savers and their social preferences. In particular, we examined not only the financial behaviour of savers but also the relationship between their behaviours, their local community and social issues. Our analysis offers an important and innovative look at the reality of our country, based on the administration of a questionnaire to a representative sample of the adult population of Trentino. The empirical evidence shows that savers are very keen on ethical investments, to the point of preferring them to their conventional equivalents.
Our analysis confirms that the intermediate age group (the under 65s) and a higher education level have a positive impact on the knowledge of crowdfunding and a propensity to save in the future. Furthermore, people are deeply interested in investing as much as possible in ethical products, that is, those that do not yield interest but that contribute to social projects that benefit the local community.
From a policy standpoint, the next step should therefore be to initiate and promote more vigorously the meeting between the demand and the supply of this type of financial products, to identify social needs and projects aimed at matching the characteristics and the key points for community strengthening. Suitable policy design interventions must be applied to build the framework necessary for this type of investments, removing or at least smoothing out any obstacles to operations to favour their progressive development.
For many reasons, it seems to us that social and welfare policies and the labour market offer significant opportunities. In fact, socially motivated local operators and public entities with a duty to protect the vulnerable members of society should facilitate the selection of shared-content social projects and enhance the opportunities for matching the demand with the offer.
Furthermore, our results also have policy implications for the financial industry, where the existence of such products can help financial advisors to accomplish their clients' financial and social preferences in a setting where household investors have a deep community commitment. Our results are based on a questionnaire administered to a representative sample of the adult population of Trentino, which, as Gambardella et al. [6] stress, "is a region with a strong sense of independence and socio-typical traits that make it a breeding ground for innovations that often anticipate and spark developments on the national scene".
Although Trento area is the region with the highest potential for social bonds, given the high level of wealth as well as the high propensity to donate compared to the rest of the country, it would be worth exploring, from a policy standpoint, the latent demand for social products in the country as a whole. This would allow us to estimate the potential of the demand for social bonds at the country level. An additional point to be further developed is the determination of the area of programme interventions where people are more willing to invest. Matching the most interesting projects for investors with the highest possible impact for the community would be the ultimate goal. All in all, our results indicate a route to promote private-public partnerships to address community problems.