In Defense of Total Repayments

By: Elwyn Berlekamp, Bernard Pech, Hari Titan

The School Board has taken to the idea that taxes can be deferred and reinvested to earn a safe return high enough to significantly discount the cost of additional interest charges. This safe return has been chosen to be between 3% and 4% based on a mix of long term Treasuries. The School Board has asserted that as a result of this "discount rate", all financing options including CIBs, interest-only hybrids and CABs cost very close to the same amount! We question this assertion on a number of grounds.

The application of a discount rate assumes the taxpayer already has something equivalent to these financial instruments or plans to get them in their brokerage account when the School Board decides on a financing scheme that defers taxes. Most taxpayers will find this a hassle to do. Piedmont issued a CAB in August 2013 (Series E) but taxpayers were not informed that the discount rate assumptions required reinvestment.

Also for an individual taxpayer the amount deferred is so low ($100 - $200 per tax bill, for the typical appraised value in Piedmont) that transaction fees getting in and out of these safe investments will eat into their returns. This taxpayer would need to make 2 transactions per year, investing the "savings" every time they pay their lower property taxes, slowly building up their investment until it is time to start cashing out and paying the deferred taxes. All this effort by the taxpayer is in stark contrast to the simplicity of the investors who lent money to the District in the first place (e.g. buying the Series E bonds). Those lenders did one transaction and got a virtually guaranteed (AA-) safe return that compounds automatically.

For most taxpayers, deferred taxes will realistically end up in higher risk investments like the stock market, or in near zero return savings accounts, or end up spending on consumer goods, travel etc. (which is effectively a 100% loss from a reinvestment POV). Reporting the total repayment costs and repayment multipliers inherently assumes a zero discount rate on unpaid taxes with no assumption on reinvestment.

Comparing repayment multipliers correctly shows the investment challenge posed to those taxpayers who do want to reinvest deferred taxes. Applying a discount rate analysis for the whole community makes unrealistic assumptions of how taxpayers will behave in the future.

[For additional context see School Board discussion around present values from hour 2:00 to 2:42 here: ]


Berlekamp received his PhD from MIT in 1964 and became Professor of Electrical Engineering, Computer Science, and Mathematics at UC Berkeley. Berlekamp also managed an algorithmic trading fund known as Medallion. Under him, the fund's one-year return to investors in 1990 was 55%. He then sold the company to Renaissance Technologies. According to the Wall Street Journal, this fund continued to provide the highest returns of any hedge fund until 2005, with Soros' Quantum Fund coming in second.

Pech received his BA in Mathematics from the University of Paris, his Physics degree from École Nationale Supérieure des Télécommunications, and his MS in Electrical Engineering and Computer Science from UC Berkeley (1970). He is now retired after a forty years career in Silicon Valley spanning product companies (, Siebel), service companies (Teknekron), and University/Government Laboratory research organizations (UC Berkeley, Lawrence Berkeley Laboratory).

Titan received his PhD in Computer Science from the University of Waterloo in 1993. Hari is a Big-Data Scientist and has focused on forecasting and risk management using diverse and large volumes of customer and economic data. Hari is the inventor of U.S. Patent #US5745654 used in the Fair Isaac credit card fraud prevention system. Hari has written extensively regarding creative financing options being presented to the Piedmont Unified School District, built a bond comparison tool and ran for the School Board in 2014.



Average: 5 (1 vote)