Geography in Online Capital Allocation: Evidence from Equity-Based Crowdfunding

Online Crowdfunding Does Not Eliminate Geographic Investment Concentration

Even though people can invest in companies anywhere online, they still tend to invest in businesses located in their own home region. This tendency creates a challenge for digital marketplaces designed to connect capital with opportunity regardless of distance. A new study from researchers at Keio University and the University of Tokyo suggests that for companies outside major metropolitan hubs, local residents remain a vital source of funding even after they have accessed all the available digital information.

The persistence of local capital

Does a digital interface truly democratize access to capital, or does it merely provide a new window onto old patterns? In traditional finance, geography acts as a heavy friction. Investors often rely on physical proximity to gather information or build trust. Digital investment platforms—like equity-based crowdfunding (ECF) sites—were intended to dissolve these barriers. They aim to reduce search costs and present a uniform array of opportunities to a global audience.

The central question the authors investigate is whether this online access actually makes investment demand "geographically neutral." If a platform works as intended, a startup in a remote prefecture (a Japanese administrative district) should attract the same interest from a distant investor as it does from a neighbor. This assumes the information provided is sufficient. However, observing only the final investment amounts doesn't tell us why this happens. Is it because local people simply see the ads first (initial salience), or do they make a conscious choice to support their neighbors after reading the fine print?

Cracks in the search-cost reduction model

Before this study, the prevailing logic suggested that digital platforms primarily solve the "discovery" problem. Much of the existing literature on online finance argues that while geography hasn't disappeared, its influence is weakened. This is because the cost of finding a good deal is lower. Previous research has noted that participants in online markets still favor nearby projects. However, these studies often focus on aggregate funding outcomes or realized investments.

The crack in this understanding is that looking only at who eventually puts money into a project makes it impossible to distinguish between different stages of the decision process. For instance, if a local investor invests in a local firm, we don't know the cause. Did they happen to click the link first? Or did they see the details and feel a specific motivation to support their home prefecture? To test this, researchers look at the "adjacent-prefecture" comparison. If the preference is just about being "close," then people in neighboring districts should also invest more frequently.

Separating discovery from decision

To move past these observational deadlocks, the authors utilize proprietary data from Fundinno, Japan’s dominant equity-based crowdfunding platform. The researchers leverage a unique capability of the platform: it records the entire sequence of a user's journey. They can track when a user is first exposed to a campaign on a listing page. They also track when a user clicks through to view the detailed campaign page (the "viewing margin") and finally, whether they commit capital (the "investment margin").

This approach allows the researchers to treat the campaign-page view as a crucial conditioning event. By analyzing the "view-to-investment" transition, they can effectively hold the information level constant. Once a user has opened the detail page, they have access to the same business plans and risk disclosures as anyone else. Any remaining preference for local companies at this stage cannot be explained away by simple discovery.

To ensure the results are robust, the authors employ a two-way fixed-effects linear probability model (LPM). This statistical method uses user fixed effects to control for time-invariant user characteristics, such as a person's permanent residence or baseline investment propensity. It also uses campaign fixed effects to control for campaign-level attributes, such as the issuer's industry or the specific terms of the offering. This controls for the inherent differences between various users and different companies.

The study further refines this investigation by comparing three distinct groups: users in the same prefecture as the issuer, users in adjacent prefectures, and "other-prefecture" users. They also split the analysis between the Tokyo Metropolitan Area (TMA)—where economic boundaries are fluid—and non-TMA regions, where prefectural borders represent more distinct social contexts.

The non-metropolitan local premium

The findings reveal a striking pattern of "same-prefecture locality" that is highly concentrated in regional areas. The authors report that for campaigns located outside the Tokyo metropolitan area, same-prefecture users are 6.7 percentage points more likely to invest after viewing a campaign page. This is a significant jump compared to the roughly 10 percent investment rate for users in other prefectures.

Crucially, this is not a smooth gradient of proximity. The authors find that the investment premium for users in adjacent prefectures is only 0.9 percentage points. This tells us that the preference isn't just for "nearby" things. It is specifically tied to being in the same prefecture. Furthermore, this effect is significantly muted in the Tokyo metropolitan area. There, the same-prefecture premium for investment is only 0.6 percentage points.

The data suggest that for regional startups, the digital platform does not create a level playing field. Instead, the "home-prefecture" effect remains a powerful driver of capital. It persists even after the investor has been fully briefed on the company's merits. The authors note that the data do not identify the specific underlying motive. It remains unclear if this is due to local information, prior ties, or a desire to support the local economy. However, they prove the preference is not merely a byproduct of browsing habits.

Implications for regional growth

These results suggest that online access and geographic neutrality are fundamentally different concepts. A digital interface can successfully expand the reach of an issuer. However, it does not necessarily neutralize the demand for local assets.

If these findings generalize, the implications for regional development are twofold. First, for issuers in remote areas, local residents remain a critical pillar of the investor base. Even in a digital world, these residents are a vital source of capital. Second, the results highlight that digital tools alone may not bridge regional gaps. Platforms might need to consider how geographic information is displayed to help manage these local interests.

The paper leaves open several questions regarding the specific psychology of the investor. Researchers still need to determine if this premium is driven by "warm glow" altruism (supporting one's home) or by a perceived reduction in information risk (knowing the local landscape better).