Mandate Score: Concept and Applications

• Mandate Score is a scalar or thresholded variable that quantifies and triggers policy obligations, enabling comparison and optimization across varied domains.
• Its applications span from evaluating open access mandates using metrics like MAND-STRENGTH and MELIBEA to assessing compliance measures in public health and organizational productivity.
• Research shows that mandate scores, whether as policy strength indices or operational KPIs, enhance predictions of compliance and overall policy effectiveness.

Taken together, the cited literature suggests that “Mandate Score” is not a standardized technical term. Instead, it denotes several distinct objects depending on domain: an explicit policy-strength variable for Green Open Access mandates, an empirically calibrated score for expected mandate effectiveness, an outcome-based mitigation index, a welfare criterion for mandate regimes, a benchmark-relative ESG constraint, a reach-triggered accountability variable, or a robustness measure of electoral control. Several papers state explicitly that they do not define a literal mandate score and instead supply the nearest formal analogue for their setting (Gargouri et al., 2012, Vincent-Lamarre et al., 2014, Cohen, 2020, Ni et al., 23 Apr 2026, Azzone et al., 2024, Khachaturov et al., 15 Jun 2025).

1. Conceptual scope and principal meanings

Across the literature, the core commonality is not a single formula but a common function: a mandate score is a scalar or thresholded quantity used to justify, compare, or trigger obligations under a mandate. In some cases it is an explicit policy-input score, as with Green Open Access mandate strength. In other cases it is an outcome score, such as mitigation performance inferred from case dynamics. In still other cases it is a regime-evaluation criterion, such as welfare under alternative information-sharing mandates, or a constraint variable, such as a benchmark-relative ESG requirement or a platform-side reach score (Gargouri et al., 2012, Cohen, 2020, Ni et al., 23 Apr 2026, Azzone et al., 2024, Khachaturov et al., 15 Jun 2025).

The literature also distinguishes sharply between mandates themselves and scores associated with mandates. A mandate may be a legal or organizational requirement, while the score may measure any of the following: formal strength of the rule, expected effectiveness, observed compliance, downstream outcomes, or robustness of the resulting collective decision. This distinction is explicit in the COVID mitigation paper, which defines a mitigation score for observed epidemic suppression rather than legal stringency, and in the enterprise “2× mandate” study, which treats mandate success as a multidimensional scorecard rather than a single scalar (Cohen, 2020, He et al., 2 Jul 2026).

A further complication is terminological. In the ISO 15531 MANDATE paper, “MANDATE” names the standard corpus rather than a score variable, and the work concerns ontology-based extraction rather than mandate quantification (Cutting-Decelle et al., 2018). In credit-risk screening, the FS-Score is presented as a possible mandate-style screen for Russian corporate credit, but the paper does not use “mandate score” as a formal term (Ivliev, 2010). This suggests that the phrase is best treated as a family resemblance concept rather than a unified construct.

2. Open Access mandate strength and effectiveness

The most explicit and mature “mandate score” literature in the corpus is in Open Access policy evaluation. One paper defines MAND-STRENGTH as the “strength of institution's deposit mandate,” classified on a scale from 1 to 12, and then relates that score to repository deposits using negative binomial regression (Gargouri et al., 2012). A second paper studies the MELIBEA score, originally a weighted composite over eight OA policy conditions, and revises it empirically to improve its predictive power for deposit behavior (Vincent-Lamarre et al., 2014).

Before the table below, two distinctions are important. First, MAND-STRENGTH is an explicitly ordinal policy-strength score. Second, MELIBEA began as a score for estimated policy strength, but the revision in the later paper turns it into a better estimator of expected mandate effectiveness.

System	Construction	Reported relation to effectiveness
MAND-STRENGTH	Ordinal categories from 1 to 12	Stronger mandates associated with higher deposit outcomes
MELIBEA	Composite of eight weighted policy conditions; later revised	Correlation with deposit rate rises from $0.18$ to $0.36$

In the MAND-STRENGTH scale, the strongest category, 12, is “immediate deposit required + linked to performance evaluation (Liège) (no waiver option).” Other reported values are 9 for immediate deposit required with no waiver, 6 for 6-month delay allowed with no waiver, 3 for 12-month delay allowed with no waiver, 3 again for rights-retention with waiver option, 2 for deposit if/when the publisher allows it, and 1 for request, recommendation, or encouragement rather than a mandate. Using ROARMAP, ROAR, Web of Knowledge normalization, and Webometrics rank, the paper reports that deposit number and deposit rate are significantly correlated with mandate strength; that MAND-STRENGTH is highly and positively correlated with deposit average, with deposit rate when mandate age exceeds 2 years, and with total deposits when mandate age exceeds 2 or 3 years; and that the strongest mandates generate deposit rates of 70%+ within 2 years of adoption, compared to an unmandated deposit rate of 20% (Gargouri et al., 2012).

The MELIBEA study is more explicitly a score-construction paper. The original formula is a weighted combination of eight conditions,

$$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$

covering mandate/request status, opt-out, version, deposit timing, embargo length, copyright, internal use, and theses. The paper reports a small but significant positive correlation of 0.18 between the original MELIBEA score and deposit percentage, identifies three especially predictive conditions—deposit timing, internal use, and opt-outs—and revises both weights and option values. The revised score gives very high value to deposit at time of acceptance, value 5 to internal use = yes, and higher value to no deposit opt-out but unconditional OA opt-out. After revision, predictive power is reported to double, from about 0.18 to about 0.36 (Vincent-Lamarre et al., 2014).

Taken together, these papers define the clearest policy-scoring interpretation of mandate score in the corpus: a mandate score is an institutional policy-design variable whose purpose is to predict or explain compliance outcomes. They also establish an important substantive result: in this literature, the most successful mandates are not merely strong in the abstract, but are strong in a very particular way—immediate, non-waivable, and linked to performance evaluation (Gargouri et al., 2012, Vincent-Lamarre et al., 2014).

3. Outcome-based and compliance-oriented operational scores

A different use of mandate-adjacent scoring appears in public health and organizational productivity. Here the score is usually not a legal-stringency index but an operational outcome indicator or compliance-linked performance measure.

The COVID mitigation paper defines a simple mitigation score

$A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$

where $\bar c_d$ is a smoothed case count. It also defines a testing-adjusted version

$$A(d)=\frac{\bar c_d/\bar p_d}{\max_{t\le d}(\bar c_t/\bar p_t)},$$

and a horizon-limited variant $A_H(d)$, with a Hamming window and recommended width $W=7$; the testing-adjusted examples repeatedly emphasize a 30-day horizon. The paper is explicit that this is not a direct score of mandates such as masking, school closure, or gathering restrictions. It is a score of observed mitigation effectiveness as reflected in case trajectories, normalized to a jurisdiction’s own historical peak (Cohen, 2020).

The face-covering paper likewise does not define a mandate score, but it operationalizes mandate effectiveness through a before/after change in average death ratio, where the average death ratio is the monthly average of daily deaths divided by the monthly average of daily cases. Counties are classified by whether this ratio decreases or increases after state mask mandate orders, using county population, median income, education level, and New York Times mask-use survey variables (“Never,” “Rarely,” “Sometimes,” “Frequently,” “Always”) as predictors. In the final binary dataset, 47 counties had a decrease and 30 an increase; the state-level changes reported are $-48\%$ for California, $-25\%$ for Washington, and $0.36$0 for Oregon; and the best reported classifier, Naive Bayes, reaches 94% test accuracy (Lafzi et al., 2021). This suggests that, in public-health settings, any practical mandate score would need to distinguish at least three components: the existence and timing of the order, behavioral adherence, and socio-economic context.

The enterprise “2× mandate” study moves the concept again. It explicitly says that it does not define a single numerical mandate score. Instead, the mandate is operationalized through a named KPI—merged pull requests per engineer per month—and evaluated through a broader longitudinal scorecard. The headline result is that per-capita throughput eventually reached $0.36$1 the pre-mandate baseline by April 2026, while the more conservative within-developer estimate carried forward by the authors is $0.36$2. The paper further reports that reviewer load roughly doubled, automated review overtook human review, and merge and revert rates held steady (He et al., 2 Jul 2026). In this literature, mandate score is best understood as a multidimensional organizational performance assessment rather than a scalar statistic.

4. Welfare criteria and mandate-regime evaluation

In mechanism design, banking compliance, and control, the nearest analogue to mandate score is often a welfare function rather than a direct score variable. The central question becomes whether a mandate improves or worsens aggregate performance once strategic response is taken into account.

The anti-money-laundering paper states explicitly that it does not define a literal mandate score. The closest formal objects are welfare under alternative regimes—especially $0.36$3—together with TVA credit accounts and incentive-compatibility conditions. The key regime definitions are: autarky,

$0.36$4

mandated full sharing,

$0.36$5

voluntary federation without incentives, and incentive-compatible federation with TVA. The central welfare-ordering proposition is that if $0.36$6, then

$0.36$7

while if $0.36$8 and competition is sufficiently intense,

$0.36$9

and there exists $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$0 such that $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$1. The calibrated simulations reported in the paper normalize welfare relative to first best and give 54% for autarky, 56% for mandatory sharing without TVA, and 87% for TVA (Ni et al., 23 Apr 2026). In this setting, a mandate score is best interpreted as a policy-performance or regime-welfare score.

The welfarist-control paper makes this interpretation explicit in general form. It treats the societal mandate as a social-cost aggregation problem and shows that, under Pareto, IIA, and pairwise continuity, social ranking can be represented by a continuous social cost function

$$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$2

such that

$$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$3

It then identifies admissible forms under different comparability assumptions, including the Rawlsian/maximin form

$$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$4

the weighted-utilitarian form

$$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$5

and the mean-plus-inequality form under full comparability. In this paper, the closest literal analogue to mandate score is exactly the scalar $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$6, or its negation if a higher-is-better convention is preferred (Hall et al., 22 Jun 2026).

These works jointly imply that in strategic or dynamic settings, mandate score is often not a primitive statistic. It is the scalar objective function against which policies are certified, compared, or optimized.

5. Threshold-triggering architectures and constraint-based mandates

A further cluster of papers uses score-like objects not primarily to evaluate mandates, but to trigger or encode them. The score determines whether an obligation applies.

In ESG asset management, the mandate is a portfolio constraint based on asset-level ESG scores $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$7. For a portfolio $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$8 and benchmark $$40\%(C1) + 10\%(C2) + 5\%(C3) + 10\%(C4) + 5\%(C5) + 10\%(C6) + 10\%(C7) + 10\%(C8),$$9, the baseline mandate is

$A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$0

meaning that the portfolio ESG score must be at least as high as the benchmark’s. In the extension with an explicit target,

$A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$1

The paper therefore treats the relevant score as the benchmark-relative ESG score, $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$2, with $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$3 as the measurable mandate target (Azzone et al., 2024).

In the social-media anonymity paper, the operative concept is a reach score. The paper proposes a three-tier regime under which obligations escalate with communicative reach. The reach score is described as “a weighted sum of followers, shares, views, etc.” and platforms are said to be able to “aggregate these signals into a rolling reach score” over “a three-month moving window” updated nightly. A single post that crosses a threshold can retroactively elevate the account. Tier 1 preserves full pseudonymity for low-reach users; Tier 2 requires private legal-identity linkage; Tier 3 imposes independent, ML-assisted fact-check review before broad amplification for specified high-risk content (Khachaturov et al., 15 Jun 2025). Here mandate score is a trigger variable rather than a performance measure.

The score-based mechanisms paper provides a general formal vocabulary for such architectures. It does not define a mandate score, but the nearest formal objects are the submitted score $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$4, the score-based decision rule

$A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$5

and, in richer mechanisms, the score recommendation rule $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$6. This framework is designed for settings with soft information and semi-hard information, where submitted scores are observable actions and may be costly to falsify (Perez-Richet et al., 2024). This suggests that many mandate scores can be understood mechanistically as payoff-relevant, thresholded evidentiary variables that govern access, approval, or review.

A looser but still relevant analogue appears in Russian corporate credit screening. The FS-Score is not called a mandate score, but the paper explicitly presents it as suitable for a mandate-style credit screen for unrated issuers. The score is

$A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$7

with $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$8, so $A(d)=\frac{\bar c_d}{\max_{t\le d}\bar c_t},$9, and it maps to broad rating-like categories. The reported in-sample discriminatory power is about 72.7% Gini AR (Ivliev, 2010). This is best interpreted as a mandate eligibility proxy rather than a mandate metric proper.

6. Electoral mandate, coherence, and scoring-rule foundations

In electoral and decision-theoretic settings, mandate score often shifts from policy design to legitimacy, robustness, or coherence.

The parliamentary-margin paper does not define a mandate score, but it gives a direct formalization of mandate robustness in preferential parliamentary elections. At seat level, the margin of victory is the smallest number of ballot-ranking modifications needed to elect a different candidate; at parliament level, the margin is the sum of the smallest relevant seat-level margins needed to alter control. In the 2015 New South Wales Legislative Assembly, the governing coalition’s loss-of-majority margin is

$\bar c_d$0

vote changes, while 22,746 vote changes would have sufficed to give ALP/CLP government and 16,349 to give a Labor/Greens coalition government (1708.00121). This is a natural electoral interpretation of mandate score as outcome stability.

The partisan-gerrymandering comparison paper is closely related. It does not use “Mandate Score,” but it studies measures that quantify whether seat share fairly reflects vote share and whether one party enjoys asymmetrical representational advantage. The paper compares fourteen metrics, including efficiency-gap variants, mean-median difference, partisan bias, specific asymmetry, declination, buffered declination, lopsided means, and equal vote weight, and concludes that the declination is “the most successful measure in terms of avoiding false positives and false negatives” on the hypothetical elections considered (Warrington, 2018). For any mandate score intended to assess whether a legislative majority is “deserved,” these measures are the nearest established benchmarks.

At a more abstract level, several papers supply foundations for when scoring can legitimately support mandates. The proper-scoring-rules paper proves that if $\bar c_d$1 is strictly proper and $\bar c_d$2, then every incoherent credence $\bar c_d$3 is strictly dominated by some coherent probability $\bar c_d$4: $\bar c_d$5 With continuity on probabilities, the same conclusion follows by corollary (Pruss, 2021). This gives a decision-theoretic sense in which a scoring-based mandate can enforce probabilistic coherence.

The score-voting paper likewise does not define mandate score, but its project score

$\bar c_d$6

is the exact scalar that determines which projects are selected, and it proves that total score functions are strategyproof iff they satisfy CCP (Cohen et al., 2022). The geometric-scoring-rules paper, finally, shows that if one requires independence of unanimous winners and unanimous losers, positional scoring rules collapse to a one-parameter geometric family,

$\bar c_d$7

with Borda, generalized plurality, and generalized antiplurality as edge cases (Kondratev et al., 2019). These papers do not define mandate score directly, but they clarify the formal desiderata—coherence, strategyproofness, and invariance under removal of extremes—that any scoring-based mandate may claim to satisfy.

The literature therefore does not support a single universal definition of mandate score. It supports a taxonomy. In policy-design settings, mandate score is typically a strength or effectiveness variable. In public health and enterprise settings, it is more often an outcome or compliance-linked performance measure. In mechanism design, finance, and platform governance, it becomes a threshold or constraint trigger. In electoral and epistemic settings, it becomes a measure of robustness, asymmetry, or coherence. The common structure is scalarization under a mandate: a score is introduced so that a rule can be ranked, enforced, optimized, or audited.