Sunday, January 21, 2018

Poway and the billion dollar capital appreciation school bond

Where does it cost $1 billion to borrow $105 million? In Poway Unified’s School District. With a pie chart showing the full cost of a $105 million capital appreciation bond being issued in a California school district to finance capital improvements, Will Carless’s local reporting turned into a national story. (Previous work on Poway’s bond financing was put together by a retired Michigan blogger, John Thurrell, who covers municipal finance).

The power of the story surely has much to do with the chart itself, which I reproduce from the original article here:

 The reason this bond is different from garden-variety school borrowing is that the district couldn’t issue a conventional bond without raising property taxes. So they “got creative.” With the help of a financial consultant they instead issued a Capital Appreciation Bond.

With a 40 year time horizon, the district doesn’t have to start making payments on it for 20 years – long after the students who enjoy the improvements have graduated. Future taxpayers will be paying bond investors about 10 times more than the initial loan.

The local taxpayer’s association calls the deal “loan-sharking,” but I also have to wonder what role property taxes and Prop 13′s limits on property tax rate hikes played in this story. Carless details how local officials figured out a way around the politically difficult choice of a tax rate hike: Proposition C. There wasn’t enough revenue coming in from the district’s $55/$1000 per assessed value levy. Proposition C asked voters if they would be willing to extend the levy for another 14 years. The idea was to collect the revenue now and start paying the bond back later. But the CAP maneuver shows that by limiting a direct and stable source of local revenues (via the property tax rate cap – which voters may modify), politicians have the incentive (and may even prefer) to pursue more exotic, and less transparent funding mechanisms.

The news story is also the power of a basic pie chart to convey information. This picture stirred up an online debate over what kinds of information citizens are given when they go to vote on bond issues. In response to the “chart-gone-viral”, the School district is defending its choice of financing while residents are “shocked” and ”appalled.”

CAPs are getting a pretty bad rap in the press. One blogger calls them, “the poor community’s way to borrow their way out of insolvency,” with the “cancer spreading” to Los Angeles Unified. While California municipalities appears to have embraced CAPs, Michigan banned them in 1994. The NYT reports that U.S. school districts issued $4 billion in CAPs last year.

August 21, 2012

Strategy and politics in the of phrasing of bond referendum

How detailed should bond referendum be? The Arlington County Board heard comments from the public on the FY 2013 capital spending plan a few weeks ago. At issue was $153 million in local GO bond referendum that will be on the ballot on November 6th. The Arlington Sun Gazette reports there are four major “bundles.”

  • $31.946 million for Metro, neighborhood traffic calming, paving and other transportation projects
  • $50.533 million for parks, including the Long Bridge Park aquatics and fitness center and parkland acquisition
  • $28.306 million for Neighborhood Conservation and other “community infrastructure” projects
  • $42.62 million for design and construction of various school projects.

At issue was the language accompanying the bond packages. The Arlington County Civic Federation contends the $45 million dedicated to the acquatics center be listed as a separate item rather than bundled under the general category of park improvements.

Scott McCaffrey writes that the County Board has been bundling bonds under thematic groupings for many years as a strategy to lessen voter opposition, an interesting claim.

How explicit does language have to be in municipal General Obligation bond offerings? States typically require GO bond debt be subject to voter approval before issuance, but how does ballot language matter to the outcome?

While not addressing the matter specifically a few related questions have been pursued in the literature. Damore, Bowler and Nicholson in their paper, “Agenda Setting by Direct Democracy: Comparing the Initiative and the Referendum” (State Politics and Policy Quaterly, forthcoming) considers if agenda setters use the referendum process to extract greater spending than the median voter desires. Some of this research indicates that voters are less likely to support state referendum for tax increases but that between 1990 and 2008, 80 percent of bond referendum received voter approval.

As to the need for particular language, there are strategies. The Government Finance Officers Association (GFOA) lists six steps governments can take to improve their chances of getting a bond approved. This includes, “measure design” or “developing ballot language that appeals to voters and clearly explains how this measure addresses the particular issue targeted by the bonds meets the needs of the community.”

I did find anecdotal evidence that politicians struggle with language on ballot questions, in an effort to strike a balance between clarity and increased likelihood of passage. The Rockford Illinois School Board appears to be hemmed-in by how it phrases bond questions. The more detailed the questions the more legally-bound the board is to spend the money as specifically approved by voters.

Speaking of language, in writing this post I was unsure if I should be using”referenda” as the plural of “referendum”. “Referenda” sounds more natural to me but “referendum” appears to be used more often.

Given the difficulty of the original Latin grammar (referendum is a “gerund” and has no plural), it turns out there is an unsettled debate over this. Either is correct according to the Irish paper The Daily Edge. I felt better knowing that even The British Parliament debated over which plural form to use back in 1998. It turns out whether one uses the Latin “referenda” or the Anglicized “referendum” is purely a matter of taste.

August 14, 2012

Where is the coercion in land use?

On Wednesday, The Wall Street Journal published an article about Denver’s light rail expansion plan. Two Cato analysts came down on different sides of the issue. Randall O’Toole, writing at Cato-at-Liberty, says that the expansion is a waste of money. He writes:

Under RTD’s latest “rethink,” transit will no longer take people from where they are to where they want to go. Instead, planners will try to coerce and entice people to live in places served by rail transit and go where those rail lines go.

The expansion comes with a steep $7.4 billion price tag, and O’Toole is likely correct that this is too much to spend; light rails across the country lose money, and the 122-mile above-ground expansion has experienced a cost overrun from $4.7 billion in 2004. The United States is notorious for unreasonably high transit construction costs compared to other countries. Additionally, the light rail is an airport connector, an often poor use of tax dollars, particularly when the airport is located far from downtown, as in Denver.

However, O’Toole’s judgment that the new plan amounts to coercion seems to be based not primarily on the light rail’s cost, but rather on zoning rules that will distinguish the new light rail stations from some of Denver’s existing light rail stations. The land around the new stations will not be dedicated to government-owned parking lots; instead developers will have the freedom to put housing or commercial uses adjacent to the stations with parking garages as far away as 1000 feet.

Timothy Lee, a Cato adjunct scholar writes at Forbes:

If the plan is to dump government-owned parking garages and instead sell the land to private developers, that’s a clear win from a free-market perspective. And if planners liberalize zoning rules to allow high-density construction that’s illegal in most suburbs, so much the better. On the other hand, if the plan is to actively subsidize or even require dense development, that is worth criticizing. But it’s important to be clear that the problem is coercive means, not the goal of providing more walkable neighborhoods.

Lee makes a key point here. The suburban style development that we see in many parts of Denver is not the free market at work, as O’Toole assumes. Rather, more dense, urban development is outlawed in many parts of Denver and cities across the country. Both O’Toole and Lee make some good points on the plan, but if a city is going to spend too much on transit, that doesn’t mean the transit should be strangled with liberty-limiting suburban zoning laws.

July 27, 2012

Is the mortgage crisis to blame for San Bernardino’s bankruptcy?

The LA Times contains a new kind of argument on why cities like Stockton and San Bernardino are in bankruptcy. To date, politicians, analysts and journalists have drawn a direct line from rising employee costs and declining revenues to municipal fiscal stress. Harold Meyerson takes another path to reach his own destination – to burnish the image of unions and  politicians. His bankruptcy diagnosis gets lost along the way.

He blames the banks. These cities went bankrupt because, “banks were peddling subprime mortgages to poorly-paid workers.” While the banks are certainly involved in the economic and fiscal train wreck he is upfront that the goal is to weave a counter-narrative, challenging the “right and center right” story of fiscal irresponsibility and overpaid public employees.

The problem with narratives (on either the right or the left) is when they cobble together related events and actors without a theoretical framework and empirical evidence. Mr. Meyerson is holding several of the puzzle pieces but then forces them together without regard to how they fit.

Some puzzle pieces he correctly identifies: a housing bubble, the role of banks, the economic fortunes of the Inland Empire, and the fiscal effects of California’s Proposition 13. What he airbrushes or ignores are the roles of Fed Policy, government lending, regulatory and land-use interventions, the short-term incentives of politicians, the hand of special interests, unions, and erroneous accounting assumptions that generated the perfect storm for a fiscal fallout in 2008.

Stockton’s troubles are plain for all to see. Steven Malanga discusses them here. The municipality’s spending spree can be traced to an overheated housing market which drove Bay Area homebuyers into Stockton in search of cheaper properties. That lead to a 20 percent population growth and a surge in property tax revenues fueling Stockton’s appetite for redevelopment. In 2003 the city borrowed for a waterfront revitalization and a 5000-seat sports arena. They bonded for pension enhancements. In total the city issued $700 million in debt.

Part of the pension deal allowed workers to retire at 50 with 90 percent of their final pay plus COLAs. To pay for this, Stockton invested some bond proceeds into CALpers on the bet it would earn more than the interest payments on that debt. They lost that bet. The housing boom – itself the creation of decades of government interventions – created the mirage of ever-increasing revenues that encouraged politicians to play fast and loose with bonds and future promises to workers.

The next claim is that defined benefit plans have been “demonized” also misses the mark. Defined benefit plans – or annuities – have been destroyed by those who champion them most loudly. Faulty government actuarial assumptions made them appear cheap to operate. That encourage politicians to offer workers (in union negotiations) increasingly generous retirement terms all while underfunding those benefits and taking risks with plan assets. This is accounting chicanery, and sadly, it was not (and still isn’t fully) recognized as such. The blame there can be pinned on the esoteric but well-documented trouble with defined benefit pension accounting. This case has been made in great technical detail by economists and practitioners.

The right salary for a public worker can really only be determined with reference to a private sector counterpart. It isn’t backed into based on area housing prices. Biggs and Richwine find public teacher salaries are on par with a private sector counterpart (in terms of SAT scores and skills). But, salary is only one component of total compensation for public sector workers. Compensation also includes (undervalued and underfunded) pension benefits and (largely unfunded) health benefits. Public sector compensation is a big and growing part of many municipal budgets. What can be said is that the cost of San Bernardino’s police and firefighters represent three-quarters of the city’s expenditures and revenues are flat.

Again, Meyerson is holding one of the right puzzle pieces: the revenue bust that followed the housing bubble. But he fails to note that it was the government-induced housing bubble and subsequent revenue boom that tempted public officials to overextend themselves. This house of cards was supported by flawed accounting and incentivized by short-term gains. This is why to make those pieces fit one needs a theory and empirics otherwise the diagnosis of San Bernardino’s and Stockton’s bankruptcy is cast aside in service of the meme. It is “politics with romance.”

What caused these two cities to tank? A host of economic and fiscal factors and scores of regulatory interventions over many decades. Some of that can be found in the accounts and CAFRs. They are no fun to comb over but they reveal choices, bargains, and tradeoffs under constraints and contain the record of the evasions and faulty assumptions of “public choosers.”

July 26, 2012

Two cities down to their bottom dollars

This week, two large cities dealt with the consequences of fiscal irresponsibility. On Monday, Scranton, PA joined the state capital of Harrisburg in the list of municipalities that have run out of money to meet their commitments. The city has a budget deficit of $17 million, leaving the Mayor Chris Doherty without enough money to pay his employees.

He made an executive decision to slash all city employees’ compensation to reflect minimum wage in the paychecks they received this week because the city now has just $5,000 in the bank. Full payroll costs the city about $1 million, he said.

On Tuesday, the City Council in San Bernardino, CA gave approval for the city to file for bankruptcy. Investigation is underway in San Bernardino to determine whether fraud was involved in the budgets of the previous 13 years that reported surpluses when in fact there were deficits each year.

In both cases, cities have gone the extreme route of nearly spending their last dollars before making abrupt and drastic decisions. While no public investigation for criminal activity is underway in Scranton, Gary Lewis, an accounting consultant closely tracking the city’s financial crisis questions whether recent federal grants to Scranton have been managed appropriately.

It appears inevitable that Scranton will follow San Bernardino into bankruptcy filings, as it’s likely impossible to raise taxes sufficiently to cover the city’s current deficits and past debts. While these two cities have few options remaining, other cities should learn from these practices to avoid the costly and painful bankruptcy process.

This difficult situation could be avoided with increased transparency in budgets and with accrual budgeting. This budgeting process requires that cities set aside money for expenses as they agree to pay for them rather than when the bills are due. Following this process would prevent policymakers from spending money in the present that will have negative consequences down the road.

July 12, 2012

« Previous Page