Public Signals

New York’s Comptroller, Tom DiNapoli has issued a draft proposal for identifying local governments at risk of fiscal stress.

An earlier report touches on local fiscal distress. And here, former Assemblyman Richard Brodsky discusses the earlier Comptroller’s report.

It’s a good idea, and given the local governments that have already gotten themselves in trouble the past several years, perhaps even overdue. Here’s an earlier post showing signs in the data years ago that Nassau County was headed for trouble.

Overall, it looks like a good proposal. And it’s certainly necessary. So kudos to Comptroller DiNapoli and his staff for taking the initiative.

Here’s how it would work.

Using local financial data and some economic indicators, the Comptroller’s staff would evaluate each local government and school district against a set of individual measures. Exceeding a particular value for each measure would produce increasing numbers of “points.” Total points would be added up for each government and district and those exceeding total point levels would be label as high, medium or no risk. While some of the details differ for different levels of local government, the direct measures for local governments (counties, cities, towns, villages) include:

1. Year-End Fund Balance (worth 8 out of the 29 points for financial indicators)
• Assigned and Unassigned Fund Balance

Example:

General Fund Only
• 3 Points = Less Than or Equal to 5% Last Fiscal Year
• 2 Points = Greater Than 5% But Less Than or Equal to 10% Last Fiscal Year
• 1 Point = Greater Than 10% But Less Than or Equal to 15% Last Fiscal Year
• 0 Points = Greater Than 15% Last Fiscal Year

Combined Funds minus General Fund

• 1 Point = Negative Result Last Fiscal Year
2. Total Fund Balance
3. Operating Deficit
4. Cash Position
• Cash Ratio
• Cash as Percent of Monthly Expenditures
5. Use of Short Term Debt
• Short-Term Debt Issuance
• Short-Term Debt Issuance Trend
6. Fixed Costs
• Personal Services and Employee Benefits as Percent of Revenues
• Debt Service as Percent of Revenues

The environmental indicators include:

1. Population – Change in Population, 1990-2010
2. Age
• Change in Median Age of Population 2000 – 2010
• Median Age of Population, 2010
3. Poverty
• Child Poverty Rate 2010
• Change in Child Poverty Rate 2000 – 2010
4. Property Value
• Change in Property Value
• Property Value per Capita
5. Employment Base
• Change in Unemployment Rate
• Unemployment Rate
• Change in Total Jobs in County
6. Intergovernmental Revenues
Reliance on State and Federal Aid
Change in State and Federal Aid
7. Constitutional Tax Limit – Tax Limit Exhausted
8. Sales Tax Revenues – Change in Local Sales Tax Receipts

School districts would be evaluated separately by means of a similar system, but with somewhat different measures. The financial indicators would be mostly similar, except that it would not include the fixed cost measures. The school district environmental indicators would differ substantially. They would include:

1. Property Value – Change in Property Value
2. Enrollment – Change in Enrollment
3. Budget Votes
• Budget votes defeats first budget vote trend
• Change in approval percentage, first budget vote
4. Graduation Rate (percentage)
5. Free or Reduced Price Lunch

I’ve not had the opportunity to ask questions of the Comptroller’s staff, who designed this system, so I’m making certain assumptions about their intentions in specific cases. But I do have some observations.

The Tradeoff Between Ease of Understanding and Throwing Away Information and the Risk of Manipulation

The design of this system provides an interesting example of a classical tradeoff in presenting more or more sophisticated information versus keeping the system’s logic transparent and readily understood by widely diverse audiences. Using a point system as the Comptroller has proposed should be easily understood by a wide variety of audiences. This is without doubt, valuable and compellingly so. However, it is not without costs.

Such systems do not use all the information available; they can equate some governments that are not really equal and could create meaningless distinctions between some that are very similar. Even within a particular indicator, information is lost. Assigning a specific point score for a range of values, discards the information within that range. For example, assume that one municipality’s fund balance was just on the edge of being categorized with a different number of points. The difference between this government and one just the other side of the point threshold would be far smaller than it and another at the other end of its category. Another example would be a local government that borrows short-term (revenue or tax anticipation notes or budget notes). If such borrowing was greater in the last fiscal year than five percent of general fund revenue, but less than 15 percent, that government and all governments within that range are assigned two points. This equates the government that borrows five percent with one that borrows 14.9 percent and it differentiates between the governments that borrow 4.9 percent and 5.0 percent.

Because point thresholds are somewhat artificial and “notchy”, such systems are susceptible to being manipulated, especially if their rules remain unchanged for lengthy periods. For example in the draft proposal, A local government that borrows for cash management purposes for three consecutive years gets three points. In contrast, one that borrows for two years, gets two points. As such borrowing is typically to smooth out cash flow, a brief delay, moving the borrowing from late in one fiscal year to early in the following fiscal year might change the point score and the outward signs of fiscal stress, but might not represent any real difference. Worse, depending on an individual government’s normal cash flow, such behavior might increase its actual cash risk.

Differentiating Between Factors that Can be Influenced or Controlled and Those that Cannot – and the Political and Policy Implications

Clearly the intent of this system is to provide early signals of stress and perhaps encourage early intervention by the State or by local officials. Such stress can result from local economic circumstances and/or local financial management. No doubt there will be instances in which those two drivers will become muddled and some local officials will be “blamed” for a stress level that they have relatively less ability to control and, conversely, some local officials will claim their stress is due to such economic circumstances rather than their own mismanagement. These are both problems, but not an excuse to not move forward. Intentional ignorance is not to be preferred merely because some will misuse or misunderstand information. Waiting for a crisis is inexcusable. So too is failing to use information that might provide early warning.

However, when the Comptroller releases reports, this problem might be partially mitigated by clearly differentiating between between those governments or school districts that are at risk primarily based on financial indicators – which can be more directly influenced or even control by local authorities and those which cannot, namely environmental and economic indicators.

Nassau County is a case in point. Nassau is under the supervision of a local financial control board, an entity created by New York State in response to a local crisis and this is not the first time. Nassau is particularly interesting for several reasons:

• Nassau’s fiscal stress is not the result of its being poor. Unlike Erie County, NY and its City of Buffalo, which have both had financial control boards in recent years, but which have lost about half their populations over the last half century, Nassau is prosperous. Indeed, it is among the most wealthiest counties in New York State. So while Erie and Nassau Counties might have similar scores in the proposed system, it would be the result of very different causes. Both might require more oversight, but the fundamental solutions to their problems are quite different.
• Moreover, there were very early signs – if anyone looked at them – that Nassau was in trouble. Starting at least as far back as the late 90’s or early in this century, the percentage of Nassau budget that went to debt service exceed that of other like counties by a large margin. Take a look (Graph – Debt Service Trends as Percent of Total Revenue – Urban NYS Counties Excluding NYC.pdf).

Did OSC Test the System?

And the Nassau case brings us to another question. Did the Comptroller’s staff, use their proposed system and historical data, to predict stress that has already occurred? If so, did it capture all of the cases that we’re already familiar with? Did it do so far enough in advance to be useful?

I don’t know the answer to this question, but it sure would be interesting to know. If they didn’t, they should have.

Why No Penalty for the Failure of Local Governments to Timely File the Required Reports and the Data Upon Which the System Would Depend?

I’ve pored over much of these data going back to 1998 and one of the most obvious problems was that much of the local financial data the Comptroller releases to the public shows gaps. Those gaps represent local governments that either did not submit their required Annual Financial Reports, they did not submit them timely, or they were of inadequate quality for the State Comptroller to release them to the public. Failing to file a financial report is one of the “twelve deadly sins” for governmental financial managers, but there are no points assigned in the proposed system for failure to file.

I did ask State staff what they do in such cases and was told that the Comptroller does not have statutory authority to penalize a local government for failure to file. They can send nasty letters, they can release reports to the press, but that’s about it. (By way of contrast, a hospital or nursing home in New York which fails to file its Medicaid cost report is very likely to be penalized financially.

The proposed system should assign at least a couple points for failure to file timely, audited reports and data and even more for repeat offenses. Otherwise, perhaps it should lead to at least a temporary, but automatic classification as high stress.

What OSC Did Not Include

The draft proposal does something we don’t often see and the Comptroller and his staff are to be commended for it. They list the measures they thought about using, but did not. And then they explain why. Good for them.

Summary

Once this system is finalized and some experience developed, there will still be questions of whether and how to standardize the State’s responses to signs of stress. New York’s past – and that of other states – is littered with examples of local governments whose fiscal stress became much worse due to delayed identification and response.

While my observations may be taken as cautionary notes, or perhaps even criticisms, this is an excellent move and, perhaps with refinements, one that should be implemented and should be build upon.

Kudos!

Tagged as: City, County, Default, Deficit, DiNapoli, Fiscal Stress, Kudos, School District, Town, Use the Damn Data, UTDD, Village

{ Comments on this entry are closed }