On 08/26/2015 AMENEH MOUSAVI filed a Tax - Other Tax lawsuit against FUTURE IRRIGATION INC. This case was filed in Los Angeles County Superior Courts, Stanley Mosk Courthouse located in Los Angeles, California. The case status is Disposed - Judgment Entered.
Disposed - Judgment Entered
Los Angeles County Superior Courts
Stanley Mosk Courthouse
Los Angeles, California
FUTURE IRRIGATION INC.
UPSTREAM CAPITAL INVESTMENTS LLC
D.I.R. DIV. OF LABOR STDS. ENFORCEMENT
YEARGIN COLE WILLIAM
1/31/2019: Assignment of Judgment
8/26/2015: REQUEST THAT CLERK ENTER JUDGMENT AND JUDGMENT ON FINAL ORDER, DECISION OR AWARD OF THE LABOR COMMISSIONER
11/17/2015: ASSIGNMENT OF JDUGMENT
2/23/2017: REASSIGNMENT OF JUDGMENT
Assignment of Judgment; Filed by Upstream Capital Investments, LLC (Assignee)Read MoreRead Less
Assignment of Judgment; Filed by Labor CommissionerRead MoreRead Less
REASSIGNMENT OF JUDGMENTRead MoreRead Less
ASSIGNMENT OF JDUGMENTRead MoreRead Less
Assignment of Judgment; Filed by AssigneeRead MoreRead Less
Request; Filed by ClerkRead MoreRead Less
Request for Entry of Default / Judgment; Filed by nullRead MoreRead Less
REQUEST THAT CLERK ENTER JUDGMENT AND JUDGMENT ON FINAL ORDER, DECISION OR AWARD OF THE LABOR COMMISSIONERRead MoreRead Less
Case Number: BS157257 Hearing Date: October 23, 2020 Dept: 82
General Motors LLC,
California New Motor Vehicle Board, Respondent
Folsom Chevrolet, Inc., Real Party in Interest
Judge Mary Strobel
Hearing: October 24, 2020
Tentative Decision on Petition for Writ of Mandate Phase II
Petitioner General Motors, LLC (“Petitioner” or “GM”) petitions for a writ of administrative mandate directing Respondent California New Motor Vehicle Board (“Board”) “to set aside and vacate its Decision dated August 13, 2018, in Protest No. PR-2483-16, and to adopt and issue a new and different decision overruling the Protest.” Board and Real Party in Interest Folsom Chevrolet, Inc. (“Folsom Chevrolet”) oppose the petition. The court held a hearing on July 30, 2020 and made certain determinations. The court continued the hearing on the remaining issues.
The court does not repeat the Background and Statutory Scheme discussion from its July 30, 2020 tentative.
Administrative Proceedings and Decision
Folsom Chevrolet filed a termination protest pursuant to Vehicle Code sections 3060 and 3066. A merits hearing was held from January 29 to February 9, 2018. GM had the burden to establish good cause. (Veh. Code § 3066(b).) On July 27, 2018, the ALJ issued a Proposed Decision sustaining Folsom Chevrolet’s protest. On August 13, 2018, the Board adopted the Proposed Decision as the Board’s Decision. (AR 1346-1438.) Board made detailed findings under section 3061 in support of its determination that GM had not shown good cause to terminate Folsom Chevrolet’s franchise. Among other findings, Board found that GM’s use of RSI as a performance metric was unreasonable both generally and as applied to this case.
On September 27, 2018, GM filed its petition for writ of administrative mandate. On November 27, 2018, Folsom Chevrolet filed an answer.
On January 27, 2020, GM filed its opening brief in support of the petition. The court has received Board’s opposition, Folsom Chevrolet’s opposition, GM’s reply, the administrative record, and the joint appendix.
On July 30, 2020, the court held a hearing on the petition for writ of mandate. The court found it was reasonable for the Board to analyze RSI as a performance metric using the section 11713.13(g)(1)(A) standards as part of its analysis of “good cause” to terminate the Folsom franchise. The court found the Board did not incorrectly apply the legal standard set forth in Section 11713(g)(1)(a) when analyzing those factors as they pertained to termination of the Folsom franchise. However, the court found that Board did not have jurisdiction to find that GM violated section 11713.13(g)(1)(A) generally, or in this specific case.
The court continued the hearing on the remaining issues in the case.
Standard of Review
The writ petition is brought pursuant to CCP section 1094.5. The pertinent issues are whether the respondent has proceeded without jurisdiction, whether there was a fair trial, and whether there was a prejudicial abuse of discretion. An abuse of discretion is established if the agency has not proceeded in the manner required by law, the decision is not supported by the findings, or the findings are not supported by the evidence. (CCP § 1094.5(b).)
The substantial evidence standard of review applies to Board’s decision on a franchise termination protest. (Kawasaki Motors Corp. v. Superior Court (2000) 85 Cal.App.4th 200, 203.) Substantial evidence is relevant evidence that a reasonable mind might accept as adequate to support a conclusion (California Youth Authority v. State Personnel Board (2002) 104 Cal. App. 4th 575, 584-85), or evidence of ponderable legal significance which is reasonable in nature, credible and of solid value. (Mohilef v. Janovici (1996) 51 Cal. App. 4th 267, 305 n. 28.) Accordingly, “[i]t is for the [agency] to weigh the preponderance of conflicting evidence, as [the court] may reverse its decision only if, based on the evidence before it, a reasonable person could not have reached the conclusion reached by it.” (McAllister v. California Coastal Com. (2008) 169 Cal.App.4th 912, 921; Sierra Club v. California Coastal Com. (1993) 12 Cal.App.4th 602, 610.)
On questions of law arising in mandate proceedings, the court exercises its independent judgment. (Christensen v. Lightbourne (2017) 15 Cal.App.5th 1239, 1251.) “In the context of review for abuse of discretion, an agency’s use of an erroneous legal standard constitutes a failure to proceed in a manner required by law.” (City of Marina v. Bd. of Trs. of the Cal. State Univ. (2006) 39 Cal.4th 341, 355.)
The petitioner seeking administrative mandamus has the burden of proof and must cite to the administrative record to support its contentions. (See Bixby v. Pierno (1971) 4 Cal. 3d 130, 143; Steele v. Los Angeles County Civil Service Commission, (1958) 166 Cal. App. 2d 129, 137; see also Alford v. Pierno
“In reviewing the agency's decision, the trial court examines the whole record and considers all relevant evidence, including evidence that detracts from the decision.” (McAllister v. California Coastal Com. (2008) 169 Cal.App.4th 912, 921.) However, “a trial court must afford a strong presumption of correctness concerning the administrative findings.” (See Fukuda v. City of Angels (1999) 20 Cal. 4th 805, 817.) When an appellant challenges “’the sufficiency of the evidence, all material evidence on the point must be set forth and not merely [its] own evidence.” (Toigo v. Town of Ross (1998) 70 Cal.App.4th 309, 317.)
Findings Not Challenged by GM
Board’s detailed decision includes approximately 189 findings of fact that span 56 pages. (AR 1363-1419.) Board’s findings address, in detail, the good cause factors set forth in section 3061. In its writ briefs, GM does not specifically challenge the vast majority of the Board’s findings. Rather, GM’s writ briefs focus predominately on a subset of Board’s findings regarding the reasonableness of the RSI performance standard.
As noted, GM bears the burden of proof under CCP section 1094.5. (Alford v. Pierno A reviewing court “will not act as counsel for either party … and will not assume the task of initiating and prosecuting a search of the record for any purpose of discovering errors not pointed out in the briefs.” (Fox v. Erickson (1950) 99 Cal.App.2d 740, 742.) The court is not a “tacit advocate” for the parties. (Quantum Cooking Concepts, Inc. v. LV Associates, Inc. (2011) 197 Cal.App.4th 927, 934; see also CRC Rule 3.1113(a); Nelson v. Avondale HOA (2009) 172 Cal.App.4th 857, 862-863 [argument waived if not supported by reasoned argument and citation to authorities]; Inyo Citizens for Better Planning v. Inyo County Board of Supervisors (2009) 180 Cal.App.4th 1, 14 [court does not serve as “backup” counsel].)
GM has failed to show a prejudicial abuse of discretion for any fact findings not specifically challenged in its writ briefs. The court concludes that those unchallenged findings are supported by substantial evidence.
Does Substantial Evidence Support Board’s Relevant Findings about GM’s Use of RSI?
GM contends that, for various reasons, Board’s findings regarding RSI were not supported by substantial evidence. (OB 13-24.)
Additional Factual Background – RSI
The administrative decision succinctly describes three terms – Area of Primary Responsibility (APR); Area of Geographic Sales and Service Advantage (AGSSA); and Retail Sales Index (RSI) – which are important to this writ petition. The following findings are not disputed:
“In the regular course of business, General Motors assigns a geographic area to each dealer, whether as an APR [Area of Primary Responsibility] or an AGSSA [Area of Geographic Sales and Service Advantage] or both with periodic updates.” (AR 1369, ¶ 63.) “APRs and AGSSAs consist of a certain number of assigned census tracts, as those tracts are defined by the U.S. Census Bureau. The collection of census tracts assigned to a dealer is principally determined by the geographical proximity of the dealership location and the population center of each tract.” (Ibid.) “AGSSAs are based primarily on proximity of contiguous census tracts to the nearest dealership ….” (AR 1370 ¶ 65.)
“The Sacramento APR is a large geographic area shared by Folsom Chevrolet with four other Chevrolet dealers. These are: 1) John L. Sullivan Chevrolet (Roseville), 2) Performance Chevrolet (Sacramento), 3) Kuni Chevrolet (Sacramento) and, 4) Maita Chevrolet (Elk Grove)…. The AGSSA assigned to Folsom Chevrolet, which is part of the APR, is specific to Folsom Chevrolet.” (AR 1370, ¶ 64.)
“Retail Sales Index or some variant of it has been used as a metric throughout the automotive industry…. General Motors' RSI metric is the ratio of dealer retail sales to expected retail dealer sales.” (AR 1376, ¶ 92.) “RSI compares the number of new retail vehicles sold by Folsom Chevrolet against the number that it was expected to sell as formulated by General Motors based on Chevrolet's statewide market penetration, and the number of new vehicles actually registered in Folsom Chevrolet's AGSSA by segments.” (AR 1377, ¶ 94.) “A RSI of 100 indicates a dealer achieved its sales expectations, i.e., state average performance.” (AR 1377, ¶ 93.)
“To determine RSI, General Motors first looks at new vehicle registrations, grouped by segments across California…. General Motors bases its calculations for RSI on California market share rather than national share.” (Id. ¶ 95.)
“As an example, General Motors looks at the vehicle segment ‘Large Pickup -Crew Cab’ total registrations for all manufacturers in an AGSSA. It does not matter where in the AGSSA the registration of the vehicle is located. Next, General Motors looks at the state average market share for Chevrolet for that segment. Then General Motors multiplies the registrations in the AGSSA by Chevrolet's California market share for that segment, for the number of sales to equal state average, i.e., expected sales, or in other words the product of that calculation is equal to the number of Chevrolet vehicles that would be registered in that AGSSA if General Motors' market share were at its state average level.” (Id. ¶ 96.)
Substantial evidence supports these general findings about RSI, including the reports of GM’s expert, Sharif Farhat (AR 3175-3260, 3454-3542) and Folsom Chevrolet’s expert, Edward Stockton (AR 2559-2699, 2701-2733), as well as testimony from the hearing (see e.g. AR 2099-2100 [Stockton], 1508 [Michael Stinson].)
GM’s Contention that RSI is Reasonable in Light of All Existing Circumstances
GM asserts various reasons that it believes that RSI is a reasonable performance metric. GM cites to evidence that RSI is “fundamental” to the auto industry, and that GM has used RSI since the late 1970s. (OB 13, citing AR 1376-77.) GM cites evidence that RSI is based on actual vehicle registration data, not projections or samples, and that RSI takes into account economic conditions, population changes, and changes in market share. (OB 13 citing AR 1900-01, 1910, 1925, 1651-52, 1744-45.) GM asserts that RSI does “a remarkable job at accounting for multiple potential causes of poor sales performance,” although it is not “perfect.” (OB 14.) In this part of its brief, GM does not identify a specific Board fact finding that GM contends is not supported by substantial evidence, which is the court’s inquiry under CCP section 1094.5.
Evidence Regarding Variation in Chevrolet Market Share: Paragraphs 99 and 100 of Board’s Decision
Board’s determination that RSI is unreasonable as applied to Folsom Chevrolet was based, in part, on evidence that Folsom Chevrolet is disadvantaged by RSI due to its location. (AR 1378 ¶ 99.) GM challenges related fact findings made by Board in paragraphs 99 and 100 of the Decision. (See OB 15-17.)
Paragraph 99. In paragraph 99, Board cited evidence that “[t]he average RSI for dealers in the state of California (less Folsom Chevrolet) is 132.6, but the average for Sacramento area dealers, excluding Folsom Chevrolet, is 97; a more than 35 -point differential….” (AR 1378, ¶ 99.)
GM contends that Board abused its discretion by relying on the “simple average” of 132 RSI to support its finding that acceptance of Chevrolet is not uniform throughout California. (OB 15.) GM cites testimony of Stockton, Folsom Chevrolet’s expert, that “the dealer body produces something very close to 100 percent RSI on average,” and “the real average RSI” is not “132 for any given dealer.” (AR 2100-01.) Rather, the 132 average is skewed by several smaller dealers with very low sales expectations and very high RSIs. (Ibid.) Mr. Stockton continued: “in all the comparisons that look at the bars of other dealers, next to Folsom, the difference between 132.6 and the real average of California dealers that’s going to be closer to and just above a hundred, that doesn’t have anything to do with what we would expect from Folsom, a dealer in a larger market. That’s just reflecting the skew.” (Ibid.) Stockton testified that the simple average of 132 RSI is “inflated by … close to 30 percent.” (Ibid.)
GM’s arguments related to paragraph 99 of the Decision ignore the basic finding made by the Board: “Acceptance of Chevrolet is not uniform throughout California.” (AR 1378, ¶ 99.) Substantial evidence supports that finding. (See Ibid., citing AR 2565 ¶ 29, 1911 at 56, 3183 ¶ 25.) Board appears to have cited a chart showing the 132.6 RSI average from 2015 to show the dispersion in California of actual sales compared to expected sales as reflected in the RSI metric. (See AR 3209 [chart].) GM’s cited testimony from Stockton, as well as this chart, support Board’s finding that acceptance of Chevrolet is not uniform throughout California. Paragraph 99 of Board’s decision is supported by substantial evidence.
Paragraph 100. In paragraph 100, Board made the following finding: “Chevrolet's performance in California is not at the same level as in the United States…. [T]he California counties that exceed the U.S. average are very close together or ‘clustered.’ Mr. Stockton presented two sets of maps, the first group attached hereto as Attachment E shows Chevrolet's market share by California county compared to the national average (registrations over/under U.S. average) and the second group attached hereto as Attachment F shows California's market share by California county (registrations over/under California average) for the years 2012 through 2016…. The first group of maps shows only a cluster of counties, generally in the Central Valley, outperform the Chevrolet national average. Folsom Chevrolet is in northern California…. This clustering shows that the variation in market share is ‘systematic,’ as opposed to being checkerboard across the state, which would mean that the variation in market share is random…. The second group of maps, counties that exceed the California average market share, also show clustering. This result indicates Chevrolet does not have ‘consistent, cohesive appeal’ across the State of California, and the variation in appeal is not random. This clustering includes generally the Central Valley, some of the central coast counties, the Inland Empire of California and a few northern California counties, but not those in the Sacramento APR. Therefore, the sales expectation for Chevrolet vehicles cannot be uniformly applied across the state…. The clustering cannot be explained by dealer performance either because if dealer performance was causing the variation, it would appear more random; there is no reason all of the strong Chevrolet dealers would decide to locate in the Central Valley, and all of the weak dealers would choose to locate in northern California….” (AR 1378-79, ¶ 100.)
The essential Board finding in paragraph 100 is that “Chevrolet does not have ‘consistent, cohesive appeal’ across the State of California, and the variation in appeal is not random.” In support, Board cited the maps discussed in paragraph 100 (see AR 2595-2604), as well as testimony from expert Stockton about the maps. (See AR 2102-04.) Stockton testified, for instance, that “the more you see clustering [in the maps], the more that tells you that there are systematic differences in how the brand is perceived.” (AR 2103.) The maps and Stockton’s testimony about the maps are substantial evidence that supports Board’s findings in paragraph 100.
GM’s arguments to the contrary with respect to paragraph 100 are not persuasive. (OB 15-17.) GM contends: “These maps do not take into account the types of vehicles purchased in each location. Chevrolet has higher market penetration in pickup truck than in car segments (compare AR 3033 with 3035), so any area with a high level of pickup truck purchases—such as the agricultural counties in the Central Valley—will appear to have higher overall Chevrolet market share than areas with more car purchases.” (OB 16.) GM does not support this argument with citation to the record. GM’s cited pages (AR 3033, 3035) do not appear related to the maps discussed by Stockton. As argued by Folsom Chevrolet, the map titles suggest that the maps take into account segment data, i.e. the types of vehicles purchased. (See Folsom Chevrolet Oppo. 17; see e.g. AR 2603 [average is “Based on Chevrolet as a Percent of Competitive by Polk Segment”].) GM does not respond to this argument in reply. Nor does GM cite any expert testimony to rebut Stockton’s testimony about the maps. In any event, even if there was contrary evidence, Stockton’s testimony and the maps are substantial evidence that “Chevrolet does not have ‘consistent, cohesive appeal’ across the State of California, and the variation in appeal is not random.” (See In re Marriage of Mix (1975) 14 Cal.3d 604, 614 [the testimony of a single witness may constitute substantial evidence].)
In its discussion of paragraph 100, GM appears to challenge certain broader findings made by Board with respect to GM’s use of RSI in paragraph 220. (OB 16-17.) Board concluded that RSI “does not consider the following: demographics in the dealer's area of responsibility; geographical and market characteristics in the dealer's area of responsibility (market competitiveness); the availability and allocation of vehicles and parts inventory; local and statewide economic circumstances; or historical sales, service, and customer service performance of the line -make within the dealer's area of responsibility, including vehicle brand preferences of consumers in the dealer's area of responsibility.” (AR 1417-18, ¶ 220.) Board’s findings in paragraphs 99 and 100, about variation in brand appeal across California, support this broad finding. Additional evidence, discussed below, also supports paragraph 220.
Board’s findings in paragraphs 99 and 100 are supported by substantial evidence.
Paragraph 223: Did Board Mistake Correlation for Causation?
GM contends that Board incorrectly found a causative effect from a regression analysis performed by expert Stockton. (See OB 17-19, discussing ¶¶ 101-103, 220, 223 of Board’s Decision; see also Reply 14-15.)
In the writ papers, GM cites the following findings made by Board about Stockton’s regression analysis. “To discern more closely whether the market variation that appears as the clustering of certain counties is based on a variable other than market segment, … Mr. Stockton used regression analysis on a census tract by census tract basis, comparing for a select five -county area surrounding Folsom Chevrolet to those outside of that area. The regression analysis indicates a) the market share component of RSI fails to take into account meaningful differences in market areas, and b) that even after taking these differences into account, the Folsom area is statistically different in terms of its acceptance of the Chevrolet brand. The factors that correlate with Chevrolet's market share elsewhere in California are different than those in the five -county area…. Mr. Stockton concluded that General Motors is incorrectly attributing sales performance to failure by Folsom Chevrolet to variations within its control, when it is due to factors outside Folsom Chevrolet's control.” (AR 1379-80, ¶ 101.)
“Controlling for the demographic variables of age, median household income, education level (25 years or older with at least a 4 -year degree), and population density and whether the dealership is in the five -county area versus the state as a whole, results in a reduction of Folsom Chevrolet's RSI requirement for 2016 by approximately 30 percent, from 1,324 expected sales to 940.” (Id. ¶ 102.)
In support of these findings, Board cited Stockton’s testimony and report. (AR 2104-05; AR 2565, ¶ 30; AR 2605-07.) In his regression analysis, Stockton compared Chevrolet registrations against demographic variables (such as average age and income) and determined that there were four demographic variables that were correlated with statistical significance for Chevrolet registrations in the state of California as a whole: Median Age, Median Household Income, % of Population with a Degree, and Population Density. (AR 2607; see AR 2104-05.) Stockton found that only two of these variables – age and % degree – were statistically significant within the “5 County Area” that encompasses Sacramento. (AR 2606, AR 2104-05, AR 2410-11.) Based on P-value results, Stockton testified that the regression analysis established it is infinitesimally unlikely that the variations by area for sales performance on the RSI metric discerned in the statistical analysis were random. (AR 2105 at 36:11-17; see also AR 2114.)
Board’s findings in paragraphs 101-102 of the Decision are supported by substantial evidence. In all material respects, Board accurately summarized Stockton’s testimony and report. GM does not show otherwise with discussion of the record. (See OB 17-18.) GM does not dispute that Stockton’s regression analysis obtained the results to which he testified. (AR 2104-05, 2606, 2114.)
GM challenges findings Board made about GM’s use of RSI which were based, in part, on Stockton’s regression analysis. (OB 17-19, discussing ¶¶ 220, 223 of Decision.) Specifically, GM challenges the following Board finding in paragraph 223:
Accounting for brand bias by controlling for demographic variables of age, income education level, and population density, and whether the dealership is in the five -county area, results in a reduction of the RSI requirement for Folsom Chevrolet by approximately 30 percent. (RT Vol. VII, 34:25-35:8) A metric that fails to account for the brand bias that the Vehicle Code requires it to account for, and which results in a sales requirement inflated by 30 percent, is not reasonable in light of all circumstances. (AR 1418-19, ¶ 223.)
GM contends that Board’s reasoning in paragraphs 223 violates the rule that “correlation does not equal causation”: “The Board has taken a correlation between several variables and applied that correlation as though it were causative. In other words, the Board asserts that because Chevrolet registrations correlate with median age within the state of California, the median age of an area causes Chevrolet registrations to rise or fall—and even worse, RSI is per se unreasonable because it does not take that ‘fact’ into account.” (OB 18 [citing cases].)
Although it is well known that “correlation does not equal causation,” GM suggests incorrectly that evidence of correlation cannot support a factual finding. “[W]here evidence of correlation itself is potentially relevant and unlikely to mislead the jury, an expert who reliably discerns this relationship can present such conclusions to the [trier of fact].” (U.S. v. Valencia (5th Cir. 2010) 600 F.3d 389, 425; see also U.S. v. W.R. Grace (9th Cir. 2007) 504 F.3d 745, 765 [holding that “the fact that a study is associational—rather than an epidemiological study intended to show causation—does not bar it from being used to inform an expert's opinion about the dangers of asbestos releases”].)
More importantly, Board’s findings do not equate correlation with causation. Board does not find age or level of education causes a person to buy or not buy a GM vehicle. Rather, Board finds there is a correlation which is a factor which should have been included in predicting sales. To use GM’s example of a correlation between sales of ice cream and drownings, it would not be inaccurate to predict that when ice creams sales are up, there will also be more drownings. This is not to say the sale of ice cream causes increased drownings. In that example, the causation factor is summertime. However, it does not matter what the causation is, the correlation is still accurate. Board could reasonably conclude that a statistical correlation between age and education level, and the sale of GM vehicles was likely to be predictive of future sales.
Folsom Chevrolet argues, correctly, that Stockton’s regression analysis and related testimony are only some of the evidence that Board relied upon to conclude that GM’s use of RSI as a performance metric, as applied to this case, is unreasonable. (Folsom Chevrolet Oppo. 23-24.) Stockton’s statistical evidence is supported by non-statistical evidence, including the maps showing the clustering of over-performing counties, the underperformance of dealer sets with similarly sized geographies to Folsom, and the underperformance of dealers in the Sacramento APR, among other evidence. (AR 3209-10, 1911 (56:11-24), 1277 (¶ 99), 2103-04, 2595-2604, 3184.) Thus, Board’s conclusion that RSI “fails to account for brand bias” is not based solely on Stockton’s regression analysis.
Stockton was qualified as an expert, and he could testify as to the weight and inferences that should be given to the correlations he found. As found by Board, “[b]y using RSI, General Motors is taking the California statewide average of Chevrolet retail sales and applying it to every dealership in California, with only one adjustment to account for only one metric, the market segment preferred in the AGSSA.” (AR 1378, ¶ 98; see also AR 1508, 2563.) As Stockton testified, the purpose of his regression analysis is to test the assumption in RSI “that the only thing that should cause a market share to vary is the types of vehicles registered.” (AR 2104 (32-33).) Stockton explained the results of the analysis as follows:
And what I find is that statistically, we’re told that to an extremely high degree of certainty, those counties [surrounding Folsom] are different from the state. So in other words, the factors that are – that correlate with Chevrolet market share elsewhere statewide are different. They’re having a different affect than what they are having in the area including Folsom.
So from a statistical perspective, … we would … reject the hypothesis that the state average market share tells use something about Folsom’s market, and that’s to a very high degree of statistical certainty.
So this is a big problem for GM’s assumption about uniform market share ….
And then the second thing we’re seeing is that if I consider that five county area, if I derive the expectation for Folsom from that area [and control for demographics], it says that the market share expectation is declining by 30 percent. (AR 2104 [emphasis added].)
Thus, Stockton testified about the correlations found by his analysis, and he explained reasonable inferences that could be made from the evidence. When combined with other non-statistical evidence summarized above, Stockton’s regression analysis and expert testimony appear to be substantial evidence that support the findings made by Board in paragraph 223. Board did not confuse correlation for causation. Rather, Board made inferences from the evidence that (1) controlling for demographics would reduce the RSI requirement for Folsom Chevrolet by approximately 30 percent; and that (2) RSI does not accurately account for brand bias as applied to Folsom Chevrolet. Nevertheless, Board has not explained how GM should take this correlation into account in setting sales expectations, or whether the other franchises in the 5 county area studied by Stockton show a similar failure to meet sales expectations based on these same correlative factors. Counsel for the Board is invited to comment at the hearing.
GM contends that “the variables appear to have minimal to no impact on RSI in isolation (AR 1380–81 ¶ 103), and a chart of dealers and their demographics in the “5 County Area” does not reveal any apparent trends. (AR 3474.)” (OB 19.) GM cites to the following Board summary of the testimony of GM’s expert, Farhat: “General Motors' counter to Mr. Stockton's regression analysis was to take each variable in isolation and review the performance to see if it showed significant deviations with respect to RSI.” (AR 1380, ¶ 103.) Board gave more weight to Stockton’s analysis, and GM does not show that Board’s weighing of the expert testimony was unreasonable. The chart cited at page 3474 is not a regression analysis. Moreover, the chart shows a substantial difference in population density between Folsom Chevrolet and the John L. Sullivan dealer, to which GM compares Folsom Chevrolet.
Subject to further argument at the hearing, the court concludes Board’s findings in paragraph 223 are supported by substantial evidence. Board did not mistake correlation for causation.
Board’s Findings that RSI was Unreliable Because Folsom Chevrolet’s AGSSA Is “Flawed”
GM contends that substantial evidence does not support Board’s finding that RSI was unreliable because “the assigned AGSSA in this case [is] flawed.” (OB 19.) Specifically, GM challenges the following parts of paragraph 221 of Board’s Decision:
[T]he General Motors RSI metric and the assigned AGSSA in this case are flawed…. As for Folsom Chevrolet's AGSSA, it was assigned an unfair AGSSA in size and distances of registrations from the dealership location, with, as noted above, required absorptions of portions of two poorly performing or underperforming terminated dealerships, the fact that it is part urban and part rural, and is an AGSSA which grew over 80 percent in registrations between 2010 and 2014. (AR 1418, ¶ 221.)
Section 11713.13(g)(1)(A)(ii) requires that the standard be reasonable in light of “[g]eographical and market characteristics in the dealer's area of responsibility.” Relatedly, in determining whether GM has shown good cause for termination, Board must consider “the amount of business transacted by the franchisee, as compared to the business available to the franchisee.” (§ 3061(a).)
The Board’s relevant findings in paragraph 221 were based on more detailed findings made earlier in the Decision, including in paragraphs 165-169. (See AR 1397-99.) Because these findings are important to paragraph 221 and GM’s writ arguments, the court quotes them at length:
“Folsom Chevrolet is located relatively close to several Chevrolet competitors, and post -Old GM's bankruptcy, is not centrally located within its current AGSSA…. General Motors increased the number of census tracts in Folsom Chevrolet's AGSSA by more than double (32 to 72) from 2010 to 2014, and the majority of the area ‘inherited’ by Folsom Chevrolet comprised geography where the prior terminated dealers had not been selling many Chevrolets.” (AR 1397, ¶ 165.)
“The new additions to Folsom Chevrolet's AGSSA resulted in two problems with regard to the requirement to meet 100 of its assigned RSI: 1) Folsom Chevrolet had to increase penetration in areas in which the two prior Chevrolet dealers had been terminated for low Chevrolet registrations and, 2) The additions to Folsom Chevrolet's AGSSA were at a greater distance from its location which resulted in the so-called ‘geographic sales and service advantage’ being flawed.” (AR 1398, ¶ 166.)
“RSI makes no allowance for the size of the AGSSA and the distance of registrations from the dealership. Analysis by both experts showed that the greater the distance of the dealership from a registration, the less likely the dealership is to capture a sales opportunity…. Mr. Farhat, General Motors' expert, looked at a composite of the other four Sacramento dealers and the percent of sales captured based on proximity from each dealership by miles and compared it to what Folsom Chevrolet was capturing from its dealership at the same distance. Within two mile ‘rings’ of each dealership, the other four dealers were capturing 39.2 percent and Folsom Chevrolet was capturing only 19 percent; within a two to four mile ring, Folsom Chevrolet captured 21 percent compared to the other's 34.7 percent. For every increase in distance from each dealership, Folsom Chevrolet captured less than the average of the other four…. Mr. Farhat's conclusion from this analysis is that Folsom Chevrolet was not effectively capturing its sales opportunity, and additionally, that the analysis confirmed the reasonableness of the RSI and ‘did its job in identifying an ineffective dealer.’” (Id. ¶ 168.)
“Mr. Farhat's sales effectiveness by distance analysis showed that the other four dealers in the Sacramento APR, which are meeting close to the 100 RSI standard at an average of 97, capture only 8.5 percent of Chevrolet registrations that are between a distance of 12 and 14 miles from their dealership, and 7.5 percent of those between 14-16 miles away…. In Folsom Chevrolet's expanded AGSSA, the next closest population center on Highway 50, Shingle Springs, is over 15 miles from Folsom Chevrolet. At 15 miles, using Mr. Farhat's data of what the other four dealers were achieving at that distance, Mr. Stockton found in his ‘ring analysis’ that Folsom Chevrolet can only expect to capture 7.5 percent of the registrations there…. If the same effectiveness by distance of the other four dealers in the Sacramento APR were applied to Folsom Chevrolet's AGSSA for 2015, the result would have generated an RSI sales expectation of 617 units within 20 miles of the dealership, still some 525 sales short of their RSI sales expectation of 1,142 units…. Mr. Stockton attributed the ability of the other dealers in the Sacramento APR to be closer to 100 because their potential customers are closer to them than Folsom Chevrolet's potential customers are to it…. This would mean that Folsom Chevrolet's inability to capture many sales beyond 20 miles is not necessarily a ‘failure’ by Folsom Chevrolet because most dealers capture only seven percent of the sales at that distance. The RSI metric is creating a sales opportunity expectation that is not based on reality.” (AR 1398-99, ¶ 169.)
Except for the last sentence in paragraph 169, GM does not appear to challenge Board’s detailed findings in paragraphs 165-169 of the Decision and Board’s summary of the “ring” analyses performed by Farhat and Stockton. These findings are supported by substantial evidence, including the expert testimonies and reports of Stockton and Farhat. (See e.g. AR 1926, 3187, 3251-3252 [Farhat’s “ring” analysis]; AR 2108-09, 2705-06, 2714 [Stockton’s response to “ring” analysis].)
GM contends Board’s interpretation of the “ring” analyses was “clearly erroneous.” According to GM, Farhat’s ring analysis “was comparative, not prescriptive” and “cannot reasonably be used to model what Folsom Chevrolet’s performance should be at specific distances.” (OB 20.) As an example, GM contends that “the ring data only extends out 20 miles, and Folsom Chevrolet’s territory is much larger than that.” (Ibid.) GM’s short arguments are not persuasive as written. GM does not cite to evidence to support these assertions. Nor does GM persuasively show that it that the fact the ring data only extends out 20 miles is material to the Board’s findings.
GM challenges Board’s finding that the “RSI metric and the assigned AGSSA in this case are flawed…. As for Folsom Chevrolet's AGSSA, it was assigned an unfair AGSSA in size and distances of registrations from the dealership location.” (AR 1418, ¶ 221.) That finding is supported by substantial evidence discussed at length in Board’s decision at paragraphs 165-169, much of which is not discussed or challenged by GM.
Board found, and GM does not dispute, that “RSI makes no allowance for the size of the AGSSA and the distance of registrations from the dealership.” (AR 1398 ¶ 168.) Based on the “ring” analyses of Farhat and Stockton, Board then found that “the RSI metric is creating a sales opportunity expectation that is not based on reality.” (AR 1398-99, ¶ 169.) This finding is also implied in Board’s conclusion, in paragraph 221, that the RSI metric and assigned AGSSA in this case are flawed.
In these findings, Board weighed the expert testimony and made inferences about the impact of the size of AGSSA on the usefulness of the RSI metric. For instance, in response to Farhat, Stockton testified the “ring” analyses showed that “how dealers capture sales is not anywhere close to how RSI derives expectations for dealers.” (AR 2108.) The testified that the “capture rate of these dealers is declining fairly substantially,” especially beyond 12 miles. (Ibid.) “[M]any of the census tracts assigned to Folsom in the AGSSA reassignment … are well beyond 12 miles…. So it’s a terrible mismatch between RSI … and how dealers actually capture sales…. And it’s … particularly relevant to Folsom because the territory added is so far away …. [E]ven according to Mr. Farhat’s composite group of dealers [this large territory] is not really conveying much opportunity.” (AR 2108; see also 2705-06, 2714.) Stockton’s expert testimony, as supplemented by other evidence discussed in paragraphs 165-169 of the Decision, appears to be substantial evidence that “the RSI metric is creating a sales opportunity expectation that is not based on reality,” including because RSI does not account for the size of Folsom Chevrolet’s AGSSA. (AR 1398-99, ¶ 169.) The court cannot reweigh the evidence.
GM contends that “[t]he model also fails to account for contrary, real-world results within Folsom Chevrolet’s own AGSSA.” (OB 20.) Specifically, GM contends that “Folsom Chevrolet’s share of Chevrolet vehicles sold in Shingle Springs remained relatively stable, with 25.3% of the registrations in 2008, 22.5% in 2009, 32.4% in 2010, and 33.7% in 2011,” even though Shingle Springs is about 15 miles away and Folsom Chevrolet could be expected to capture about 7.5 percent of registrations there. (OB 20, citing 4317-21.) GM does not cite testimony explaining the chart at pages 4317-20. Nor does GM provide context for this data to show that it undermines findings from the more recent “ring” analyses. (See AR 2714, 3252.) Nonetheless, Respondent and Real Party in Interest do not specifically respond to this argument in their briefing. They are invited to respond to this argument at the hearing.
Finally, GM contends that “even if Folsom Chevrolet’s AGSSA were found to be unreasonably defined, the sales expectations it generated—the only relevant effect from the AGSSA in RSI—were quite reasonable.” (OB 21.) GM cites evidence that the other four dealers in Sacramento area performed at an average of 97.1 RSI. (Ibid.) GM misstates the Board finding. Board did not find that AGSSA was “unreasonably defined.” Rather, it is the expansion of the AGSSA and the distance between Folsom Chevrolet and the registrations measured by RSI that formed the basis for the Board’s conclusion. Moreover, other evidence supports Board’s finding. GM’s expert, Farhat, included in his report that the average RSI among approximately 10 dealers with the closest AGSSA size to Folsom in California, excluding Folsom, is 72. (AR 3184, AR 3223; AR 1376, ¶¶ 90-91.) Omitting the best-performing dealer in the Sacramento area, John Sullivan Chevrolet, substantially reduces the average RSI of the other Sacramento dealers to around 80. (AR 3217, 3210.) GM also does not address the evidence that each dealer in Sacramento has a unique AGSSA. (AR 1370, ¶ 64.)
Subject to further argument at the hearing, the court finds substantial evidence supports Board’s findings in paragraphs 169 and 221 that changes in Folsom Chevrolet’s AGSSA made RSI an unreliable metric as applied in this case.
Board’s Findings About Insufficient Inventory Allocation; Fleet Sales
GM contends that Board found that RSI “is unreasonable due to insufficient inventory allocation,” and that this finding is not supported by substantial evidence. (OB 21.)
In opposition, Board argues, correctly, that the findings about inventory are not referenced in Board’s discussion of good cause factors under section 3061. Nor did Board expressly incorporate its inventory findings into its discussion of RSI under section 11713.13(g)(1)(A). (See Board Oppo. 20-21; AR 1397-1419.) In reply, GM responds that inventory “was a major issue” in the case, and inventory availability is a factor under section 11713.13(g). (Reply 18, fn. 12.) Contrary to the opening brief, GM does not show that Board found that RSI, as applied to this case, was unreasonable due to insufficient inventory allocation. In that respect, GM incorrectly frames its arguments about inventory. Nonetheless, as GM points out, inventory availability is a factor under section 11713.13(g) and could also be relevant to the first good cause factor under section 3061. Thus, Board’s detailed findings about inventory may have some relevance to its findings about RSI and good cause.
On the merits, GM contends that it “presented a chart comparing vehicles available to Folsom Chevrolet during 2015 against the number of each vehicle model Folsom Chevrolet was expected to sell.” (OB 22.) GM contends that “Board chose to ignore every vehicle model where Folsom Chevrolet received more vehicles than it needed—the overwhelming majority—and count up only those vehicle models where the dealership was allocated—based upon its low sales rate—fewer vehicles than it needed to hit 100 RSI.” (OB 23; see AR 4129 [chart]; AR 1392-94 [Board inventory findings]; see also AR 1710-11 [Muiter testimony about Exhibit 277].)
Board expressly rejected GM’s interpretation of the chart submitted as Exhibit 277. Board made the following relevant findings, among others, with citation to evidence:
146. Certain models sell well, and certain models do not sell well. In many instances, clearing out slow selling models by Folsom Chevrolet did not prompt General Motors to provide Folsom Chevrolet with units that are high in demand; it just prompted General Motors to allocate more of the slow -selling units. This occurred with the small sub -compact Spark, which is not a big seller in the Folsom or greater Sacramento area. (RT Vol. VII, 213:24-215:20) Folsom Chevrolet sold 10, and General Motors for the next month requested Folsom Chevrolet take multiples of that. (RT Vol. VII, 213:24-215:20) Mr. Muiter's chart indicated that Folsom missed out on 21 Spark units because it did not request additional vehicles over its allocation. (Exh. R-277)
147. …. Mr. Muiter's chart indicates that Folsom Chevrolet should have accepted 49 additional units of Chevrolet Malibu vehicles in 2015. (Exh. R-277) So out of a total of 249 vehicles, General Motors believes Folsom Chevrolet should have accepted or had the opportunity to request 70 more Spark and Malibu vehicles. (Id.) For 2015, the Spark vehicle achieved only 5.25 percent of competitive registrations in California and the Malibu achieved only 2.31 percent. In comparison, the Chevrolet Camaro was at 30.04 percent, the Suburban at 32.75 percent and the Silverado at 29.3 percent. (See Exh. P-185-126)
149. …. Protestant's expert Mr. Stockton, pointed out that Mr. Farhat's analysis only compared the inventory Folsom Chevrolet had to actual sales, which does not reflect whether Folsom Chevrolet had adequate inventory to sell more vehicles (i.e., achieve a higher sales rate) or to reach 100 RSI. (RT Vol. VII, 71:1-15, 199:21-202:12) Mr. Farhat failed to evaluate whether Folsom Chevrolet had enough inventory to support the sales rate needed to reach 100 RSI. (Exh. P-186-6 1119; RT Vol. VII, 71:16-21, 199:21-202:12)
In its writ briefs, GM does not sufficiently discuss the evidence cited by Board in these findings re. inventory. When an appellant challenges “’the sufficiency of the evidence, all material evidence on the point must be set forth and not merely [its] own evidence.” (Toigo v. Town of Ross (1998) 70 Cal.App.4th 309, 317.) GM does not show, with discussion of all material evidence, that Board’s inventory findings are not supported by substantial evidence. Moreover, Board did not find that insufficient inventory allocation made RSI unreliable as a performance metric. Thus, in that regard, GM’s arguments misconstrue the Board’s findings.
As part of its discussion of inventory allocation, GM contends that “[t]o the extent Folsom Chevrolet suffered from inventory shortages or imbalances, the Board’s own findings show it was due entirely to the dealership’s decision to use its inventory to make lucrative fleet sales.” (OB 23; see e.g. AR 1385-87, 1402-03 [Board findings about fleet sales].) In the opening brief, GM does not dispute that Folsom Chevrolet’s substantial fleet sales business was consistent with the terms of the Dealer Agreement. (OB 23.) GM does not contend that Board’s detailed findings about fleet sales are not supported by substantial evidence. Contrary to GM’s position, these findings do not suggest that any issues with inventory shortages or underperforming RSI were caused by Folsom Chevrolet’s fleet sales.
Rather, Board noted that RSI only considered Folsom Chevrolet’s nonfleet sales to determine Folsom Chevrolet’s sales effectiveness. (See AR 1402 ¶ 176.) Folsom Chevrolet’s fleet sales were a significant portion of its business. (AR 1387 ¶¶ 130-133; AR 1389 ¶ 136.) Board found GM’s failure to account for fleet sales problematic because the parties’ franchise made no such distinction in its recitation of dealer sales obligations. (AR 1402, ¶ 177.) Thus, GM’s practice in excluding such sales from its determination of Folsom’s sales effectiveness meant that GM could not show that Folsom was inadequately transacting business as compared to that available to it. (AR 1403 ¶ 178.) Fleet sales also appear to have some relevance to Board’s finding that RSI was an unreliable performance metric, as applied to this case. GM does not show, with discussion of the record, that these findings are not supported by substantial evidence.
The Beck Decision
In its analysis of the reasonableness of RSI under section 11713.13(g)(1)(A), Board relied in part on a New York appellate decision, Beck Chevrolet Co., Inc. v. General Motors LLC (2016) 27 N.Y. 3d 379. (See AR 1416-17.) In a footnote, GM argues that Beck is not like this case because of “sharp and measurable differences in brand popularity between upstate (rural) New York and the downstate region near New York City.” (OB 16, fn. 5.) Board did not rest its analysis solely on Beck; Board went on to make its own factual determination that RSI is unreasonable based on the factors set forth in section 11713.13(g)(1)(A). In any event, Board’s analogy to Beck is supported by the record here. For instance, Beck concluded that “those dealers, like Beck, who service an assigned area in which Chevrolet is less popular are disadvantaged when measured against dealers in other parts of the state in which the Chevrolet brand is stronger and facilitates dealer sales performance.” (Beck, supra at 391.) Similarly here, there is substantial evidence that brand bias, demographics, and an enlarged AGSSA all negatively impacted Folsom Chevrolet’s RSI.
Based on the foregoing, Board’s findings related to RSI, including the reasonableness of RSI as applied to Folsom Chevrolet under section 11713.13(g)(1)(A), are supported by substantial evidence.
Board’s Application of the Good Cause Factors under Vehicle Code Section 3061
In its Decision, Board made detailed findings in support of its conclusion that the good cause factors in section 3061 weighed against termination. (See AR 1397-1419.) In its opening brief, GM devotes two short paragraphs to these findings, and incorporates arguments in the petition. (OB 24-25.) GM’s conclusory arguments are insufficient to satisfy its burden of proof under CCP section 1094.5 to show a prejudicial abuse of discretion. (Alford v. Pierno CRC 3.1113(a); Nelson v. Avondale HOA (2009) 172 Cal.App.4th 857, 862-863 [argument waived if not supported by reasoned argument and citation to authorities].) It is wholly inadequate for GM to assert, without any analysis or citation to the record, that findings made by Board are “one-sided” or “unsupported.” The court concludes that all of these findings are supported by substantial evidence.
GM also contends, by reference to the petition, “the Board invented its own standard for measuring fleet sales that fails to comply with the Vehicle Code because it ignores the statutory requirement that the business available to a dealer—i.e., potential fleet sales—also be considered.” (OB 24-25, citing Pet. ¶¶ 49-50.) GM provides no citation to the record in support of this terse argument. GM has the burden of proof to show good cause for the termination. Folsom Chevrolet’s fleet sales appear to be relevant to both to the reliability of RSI as a performance metric, and the adequacy of Folsom’s vehicle sales as compared to its sales opportunities. Board concluded: “As it has been concluded in this case that ‘sales’ includes both retail and fleet, General Motors did not meet its burden of proving that the ‘amount of business transacted’ by Folsom Chevrolet, ‘compared to the business available’ to it was inadequate.” (AR 1403, ¶ 178.) Board did not abuse its discretion in concluding that, when fleet sales are considered, GM did not meet its burden of proof of showing that Folsom Chevrolet’s sales were inadequate under section 3061(a).
GM’s Improper Incorporation by Reference of Constitutional and Other Arguments Made in Petition
In the last paragraph of its brief, GM incorporates by reference a host of constitutional arguments from the petition. (OB 25.) As noted above, GM’s incorporation by reference of arguments in the petition is improper because it exceeds applicable page limits. (See Cal. Rule of Court, Rule 3.1113.) The court rejects the arguments made in this part of GMs’ brief, as there is insufficient legal analysis. (Nelson v. Avondale HOA (2009) 172 Cal.App.4th 857, 862-863 [argument waived if not supported by reasoned argument and citation to authorities].)
The court invites further argument regarding specified issues listed above. Subject to this argument, the court tentatively DENIES the petition.
 It is unclear from the opening brief whether GM challenges the following finding in paragraph 220: “General Motors' market share is sensitive to demographic differences in the California buying populations. (Exh. P-185-8 ¶ 29).” (AR 1417, ¶ 220; see OB 17.) This finding was based on paragraph 29 of Stockton’s report, as well as the maps discussed above, which are substantial evidence.
 GM violates applicable page limits for GM by incorporating legal arguments made in the petition. (See Cal. Rule of Court, Rule 3.1113; see OB 20, citing Pet. ¶¶ 66-77.) The court considers the arguments in the brief, but disregards the arguments that are improperly incorporated by reference.
 To the extent GM argues otherwise in reply, the court finds its interpretation of the Dealer Agreement unpersuasive. (Reply 19, fn. 14.) Also, this argument was improperly made in reply.
 As noted above, GM’s incorporation by reference of arguments in the petition is improper because it exceeds applicable page limits. The court considers the arguments made in the brief.
Get Deeper Insights on Court Cases