Publications

In 1997, Japan’s banks were in crisis due to hundreds of billions of dollars of nonperforming real estate loans. In response, the government performed three rounds of capital injections in 1998, 1999, and the early 2000s. The capital injection of 1999, authorized by the Prompt Recapitalization Act, made as much as ¥25 trillion ($208 billion) available to financial institutions that applied, regardless of their capitalization. By the end of the injection window, 32 banks and trusts applied for and received ¥8.6 trillion ($71.6 billion) total in preferred shares and subordinated debts. The Act required banks to submit and adhere to restructuring plans in order to receive capital, leading to a series of mergers and acquisitions. However, differing accounting methodologies, evergreening, and double gearing allowed for systemic undercapitalization even with injections intended to help institutions meet reserve requirements.

In 1990, the asset-pricing bubble in Japan peaked and began a steady decline. Over the next seven years, a series of bank failures induced the Japanese government to introduce the first of a series of capital injections in 1998, 1999, and 2004. The capital injection of 1998, authorized by the Financial Functions Stabilization Act, made ¥13 trillion ($103 billion) available to financial institutions that applied. By the end of the injection window, 21 banks and trusts applied for and received ¥1.8 trillion ($13.5 billion) in subordinated debt and loans and preferred shares. While there were no limits on compensation for management, the Act restricted dividend payments and required banks to submit restructuring plans. However, lack of oversight over bank balance sheets to pursue risk-based injection strategies, regulatory forbearance, and banks’ application for capitalization below balance sheet needs prevented complete recapitalization of the banks and led to a second recapitalization scheme one year later.

Following the Japanese Financial Crisis in 1997, the non-performing loan problem persisted in the regional banking system, with regional banks accounting for half of all nonperforming loans by 2004. Following the capital injections of 1998 and 1999, the government created capital injection legislation to address the non-performing loan problem. Through this legislation, entitled the Act on Strengthening Financial Functions, made available ¥2 trillion ($18 billion), later expanded to ¥12 trillion ($113 billion) to recapitalize regional banking institutions. To date, 30 financial institutions have applied and received ¥674 billion ($7.3 billion) in capital injection through preferred shares, subordinated loans and debt, preferred investments, and trust beneficiary rights. Approximately ¥200 billion in capital injection has been recovered to date.

The Indonesian government implemented a joint recapitalization program in 1999 to aid some of its private banks struggling with the effects of the Asian Economic Crisis. Nine banks were eligible, and seven ultimately participated. The program was voluntary; in order to participate, bank managers had to pass a test proving that they were competent enough to run their bank, and create a three-year plan for the bank’s operations subject to independent assessment. All of the bank participants were able to return to the 4% minimum CAR requirement by the end of the program.

Conference Presentations

Ongoing or Old Projects

Honor's Thesis: FXR Regimes for Asia?

Currently, countries have two clear options when looking to anchor exchange rates to curtail macroeconomic instability: either anchoring exchange rates to the US Dollar, or anchoring exchange rates to the Euro. In either scenario, the model of exchange rate choice by Alesina and Barro posits these currencies act as imperfect solutions. The US and the European Union produce different goods and services and thus have completely different economies than the developing countries that currently anchor their exchange rates, which leads to differing monetary policy needs—policy that is controlled by the anchor country. Using a panel data set spanning 64 years of monetary policy and 179 countries, I predict the benefit to prospective anchored exchange rate regimes. In this paper, I look at the implications of the rise of the Chinese, Indian, and South Korean economies as a previously unnamed outside choice for exchange rate anchors. I also assess whether capital increases as a result of joining a fixed exchange rate regime. Examining co-movements between prices and outputs, I find that the countries that would benefit the most from anchoring exchange rates to China or India are countries from Eastern Europe: the Czech Republic, Slovakia, Croatia, and perhaps Macedonia. Currently no countries may benefit from anchoring exchange rates to South Korea. Additionally, I find that fixed exchange rate regimes cause a significant increase in capital inside a client country.

Toxicity in online forums

Something I’ve found particularly interesting recently is how communities on the social networking site, Reddit, can be quarantined. Being quarantined on reddit indicates that the contents of a forum (or ‘subreddit'), while not breaking site rules on content, can nevertheless be highly offensive and upsetting or promote hoaxes. These communities then are placed under ‘quarantine,’ meaning that Reddit users must explicitly opt-in to view this content, instead of the traditional opt-out process of the social media site. Knowing this, I had to ask: can we use sentiment analysis to see the differences in language across the quarantine line?

Market for Resident Assistants

Utilizing auction methodology might be the key to maximizing satisfaction simultaneous mass hiring scenarios.

Airlines Supply Networks

Using SNL data, I looked at the top suppliers of the top 10 airlines around the world, to see how concentrated suppliers were in air supply chains. Next steps would also be looking at other customers for air suppliers (e.g. Rolls Royce has automobiles they sell as well). You can see the results here, which I coded and graphed in R, then exported to an HTML embedding that may or may not work.