Last year’s state budget called for a revised approach to distributing new growth in California’s community colleges. Designed to address problems in the old formula, the new legislation bases funding for growth on two main factors: (1) unmet need within the geographic boundaries of a district, and (2) the effectiveness of a district’s colleges in serving people living in high-need neighborhoods anywhere in the state. The first portion is to take effect for distributing growth this fall, and the second portion the following fall.

The legislature left the details of the formula up to the system office, the California Community Colleges Chancellor’s Office (CCCCO). A draft approach is now available on the system website. Unfortunately, the draft formula has a lot of problems that need to be fixed if the most needy neighborhoods and populations in the state are to be better served:

**1.** **Using two conflicting participation rate measures**. The legislation, SB 860, asked for the growth funds to be distributed based on disparities in participation rates, with current enrollment as the numerator and people-without-degrees (and other need measures like poverty) as the denominator. The CCCCO draft, however, attempts to meld that need-based approach with a *different* version of a participation rate not based on need (enrollment divided by total population). In many instances the CCCCO approach directly contradicts the SB 860 approach.

Here’s the problem. Some districts have low participation (using the CCCCO definition) because they are already highly educated. The CCCCO participation rate sends more money in their direction while denying it to the needy communities where participation is “high” only because they are extremely under-educated. For example, the West Hills district in the Central Valley is very poor and under-educated, while West Kern is the opposite. Yet because the CCCCO formula includes even those people who already have college degrees in the denominator, both districts are considered over-funded compared to the state average. In contrast, a need-based measure shows current enrollment in West Hills is *not* high because it has enormous need. Similarly, the flawed CCCCO participation rate considers the much-underserved and under-educated San Bernardino district as being equally well-served as the highly-educated Marin County just north of San Francisco.

Think about the analogous task of distributing a limited number of vaccines. City A has 50,000 residents of whom 20% are not inoculated, while City B has 100,000 residents of whom 10% are not inoculated. There are 2000 vaccines available to split between them. A need-based approach would consider the 10,000 people in each city who are not inoculated, giving each city 1000 vaccines. The CCCCO approach, on the other hand, looks at the whole population rather than need. Seeing that City A has enough vaccine for 1/50 of its population while City B has enough for only 1/100 of its population, it would send twice as much vaccine to City B. And if everyone in City B was already inoculated, the CCCCO participation-rate measure would send all of the vaccine there, where it is not needed at all.

The people of Marin need more vaccines than San Bernardino, but they do not need more community college seats than do the people of San Bernardino.

**2. Comparing raw percentage point differences across measures**. District C has an unemployment rate of 5% and a bachelor’s degree attainment rate of 30%, while District D’s rates are 8% and 33%, respectively. With such a high unemployment rate, District D is far worse off and in greater need of more access to education and training; the difference in educational levels is not large. Yet the CCCCO approach treats these two districts as equal because there is a three percentage point difference in each rate, canceling each other out. There are various ways to address this. For example, one can take the range of rates into consideration before indexing the differences, and/or consider the amount above an acceptable or target rate (e.g. measure excess unemployment above 5%).

**3. Rounding negatives up to 1 and big numbers down to 10**. The CCCCO creates a range of negative and positive values by subtracting from the average, and then pulls a Lake Wobegon by declaring all districts above average in need, eliminating all of the negatives and cutting back the highs. By “capping the difference between 1 and 10,” the CCCCO formula pretends that Los Angeles’s unemployment rate of 9.7% indicates the same amount of need as Santa Barbara’s much lower rate of 6.6%. A different approach to indexing, as described in #2 above, would regularize different measures. (To the extent that the reason for this step was to establish minimums and maximums, the statutory language already gives districts a minimum level of growth and caps the maximum, so there is no need for the guts of the formula to also add minimums and maximum.)

An analogy. A special medicine is going to be given to a group of people based on changes in weight during college. The amount of change in the group ranged from a loss of 8 pounds to a gain of 22. To determine the amount of the drug they should get, the people who *lost* weight are assumed to have *gained* one pound, and everyone who gained 11-22 pounds is assumed to have gained just 10 pounds. (No, it does not make any sense).

**4. Applying the result to base funding**. The CCCCO formula does all of the gymnastics described above and then applies the result as a percentage increase to each district’s current funding allocation. By doing so, a district that is already well-funded gets a larger increase in enrollment than one with equal need that has less healthy current enrollment. (Imagine allocating growth to a tall child and a short child based on a percentage of their current height, year after year. The gap would get larger and larger). The result is that the latter district can never catch up: over the years the disparity only expands. The effects of this perverse approach can be seen in the CCCCO spreadsheet: Glendale gets the same growth as Sierra, even though Sierra has more than twice as many adults without degrees and a much lower base rate of participation (whichever way you measure it). Instead, the growth allocation should be based on relative need (i.e. if Sierra has twice as much need as Glendale it gets twice as much of the state’s total growth, the district’s “proportionate share of the statewide need,” as described in SB 860).

Some have argued that the measures need to be constrained because the needy districts are not equipped or do not want to grow as much as they need to. That is one reason for the minimum and maximum in SB 860 which are applied as a percentage of base funding; there is no need to also artificially deflate measured need. If our colleges are not reaching needy communities, it is critical for the CCCCO to lead a process to identify barriers and to find solutions for addressing the needs. Helpfully, the second part of the formula invites the eager-to-grow districts to play a role even if they do not have a lot of need within their district boundaries.

**5. Failing to develop or even discuss the second part of the formula**. By developing the unmet-need portion of the formula without any discussion of the effectiveness-in-serving-the-poor piece of the formula, the CCCCO is giving districts the wrong impression about their own trajectory of future growth. In particular, wealthier or better-managed districts have no advance notice that they can increase their future growth allocation by showing good outcomes for students from high-need neighborhoods inside or outside of their district boundaries. Districts that are poised to grow (for whatever reason) are obsessing over the CCCCO draft formula and trying to game it to their advantage, not realizing the opportunities offered by the second part of the formula starting in 2016-17.

The need data have been available for more than a year, and California Competes alerted the CCCCO and its Board of Governors to the flaws in the draft formula more than two months ago. So far, however, there is no sign of a revision, and time is wasting away: community college districts need to know soon what their funding allocations will be for the coming academic year.