Background
In Part 1 of this study, we examined ongoing, publicly reported art and culture funding streams in Metropolitan Detroit and Metropolitan Twin Cities and aggregated the data to compare overall art investment strategies in these metro areas.
In Part 2 of this study, we mapped these overall art investment strategies to their impact on the art and culture ecosystems in these Metro areas. In particular, we examined why the Twin Cities’ more comprehensive funding strategy outperformed Metro Detroit’s.
In this. third part of the study, we use Principal Component Analysis to reduce the dimensionality of the data collected in part 1 and map the expected effectiveness of different art investment strategies onto a single axis.
The funding streams are characterized and compared using five variables:
- Where is decision-making located (in the ecosystem)?
- How flexible is the funding (what % could be distributed differently in the next funding cycle)?
- How spatially distributed is the funding?
- How is the funding distributed by organization size?
- How is the funding distributed between project support and operational support?
These variables are all normalized to the range 0-100 to enable a comparison of funding strategies independently of funding size. In addition to the aggregated data, the following major investment strategies were studied: MN State Arts Board (MN Legacy) funding, Michigan Arts and Culture Council (MACC) funding, the Ralph C. Wilson, Jr. Foundation Arts & Culture Initiative (RCWJF), Fred A. and Barbara M. Erb Family Foundation art and culture funding (Erb FF), the Knight Art and Tech Program, and the DIA Millage. This is not a definitive study of all funding streams, but it provides sufficient data to look directionally at the likely effectiveness of different major art investment strategies.
Principal Component Analysis is a multivariate statistical technique that uses correlation among the variables to reduce dimensionality. The more correlated the variables, the fewer Principal Components need to be studied. For the chosen dataset, ~94% of the variance across the five variables is explained by the first principal component (PC1), indicating that only PC1 needs to be considered.
Results and Analysis
The ordinates of the different investment strategies along the vector PC1 are shown in the diagram below. The relative positions of the aggregate investment strategies indicate the direction of greater or less predicted impact (with the Twin Cities considered more impactful)
The investment strategies are divided into three groups (RYG) as shown in the diagram above. Analysis on the reasons for the relative position of the different funding streams are shown in the following table.
Concluding Remarks
This study (although relatively small-scale) shows the value of looking at art and culture funding at the ecosystem (rather than organizational) level, and through an objective (rather than subjective) lens. This allows the analyst to understand causal effects and hence create actionable intelligence. Ultimately, these insights can guide funders to understand the relative successes and failures of the current (as-is) funding landscape and design more impactful future (to-be) art investment strategies. Although it focuses on a benchmark of art funding in Metro Detroit in comparison to the Metro Twin Cities, the methodology could potentially be applied to any metropolitan area.
Copyright Essay’d 2026