We estimate the causal effect of partisan media on voter behavior by exploiting a natural experiment in which Sky News Australia – a conservative 24-hour news channel – became freely available to 8 million people. Using comprehensive national polling-station-level electoral data and broadcast signal strength variation, we implement a continuous treatment difference-in-differences design comparing voting patterns in areas with and without access to the channel before and after its September 2018 free-to-air launch. We find that exposure to partisan media increased the conservative party's lead by 1.5 percentage points in Australia's 2019 federal election. These findings shed light on the role of partisan media in shaping electoral outcomes by influencing swing voters, not just by affecting who turns out to vote.

Working Paper (IZA)

We employ a regression discontinuity design to estimate the causal impact of R&D tax incentives, exploiting a feature of the Australian policy which provides a higher level of support for companies with a turnover below A$20m. We estimate the policy generates an additional $1.61 R&D per dollar of tax revenue foregone. To understand the economic impact of this spending, we estimate the effect of own and external R&D stock on firm productivity, then use the estimated marginal product coefficients to calculate the implied economy wide marginal social return. Our results imply that $1 of eligible R&D generates an additional $3.10 via higher economy-wide production. Putting these together, we arrive at a net present value of the economy-wide benefit of the R&D tax incentive policy of $4.99 per dollar of revenue forgone.

Working Paper (SSRN)

This study examines the role of Victorian Government trade missions in enhancing export performance among firms, with a focus on regional targeting within high-growth markets, primarily within Asia. Trade missions are intended to mitigate the challenges of entering foreign markets by facilitating connections, reducing barriers, and providing strategic support for firms. Building on existing research, this analysis uniquely considers the impact of targeted trade missions by region, using a matched sample of firms and a stacked Difference-in-Differences (DID) methodology to isolate the causal effects on export outcomes. Results indicate that, for firms with prior export experience, participation in trade missions significantly increases export volumes in new markets within the targeted regions, suggesting that missions enable deeper market penetration. In contrast, for firms without recent export experience, the impact is more limited to initial market entry rather than growth in export volumes. These findings highlight the value of trade missions as an export promotion tool, especially when designed with firm-specific needs and regional market characteristics in mind.

Coming Soon

From 1797 to 1827, the Macarthur family specialised, collaborated collectively and persisted to create Australia's first producer-driven GVC in fine-wool. Product, process and logistical innovation were all necessary for success. In this paper, we use an in-depth longitudinal case study to demonstrate how these innovations were enabled as a result of the unique qualities afforded to family businesses. Long-term focus on value creation through sustained collective and collaborative innovation by geographically separated family members across multiple fronts was required to take advantage of the opportunity arising from the changing context of the Industrial Revolution in England, and leave a legacy.

Working Paper

The measurement of firms pursing new goods producing business models, often called fabless manufacturing or own-brand importer-marketers, has centered on industry-specific micro measures that are limited by cost, confidentiality, coverage or geographic granularity. Most often these firms that adopt these factoryless product-focused models are unhelpfully classified as wholesale traders. We introduce a new internationally linked trademark database (TM-Link) and show that variation in international trademarking activity, after controlling for exports and outward FDI, conforms well to product complexity measures and existing evidence on the growth of both fabless manufacturing and production fragmentation (by State and by industry). We argue that since companies trademark in markets where they want control over product management, design and distribution, domestic and international trademark data can shed light on the firms and locations that are engaged in these new business models.

Working Paper

Data: TM Link

Abstract The Great Depression in the United States was a unique period in terms of the movement in real estate prices. To date, only Shiller (2005) has previously attempted to provide an annual national-level housing price index. Yet previous work has suggested that potential data issues stemming from combining data constructed from a housing survey intended to examine long-term trends rather than annual movements with the need to splice additional data from a small sample of cities beginning in 1934. This paper uses a newly constructed data set from 106 cities in the Home Owners' Loan Corporation City Survey to create a new national-level housing price index from 1929 through 1940. That data set is unique in that it has housing and neighborhood characteristics for over 6,000 neighborhoods with housing prices for three separate, non-consecutive years in each neighborhood. We construct constant-relative-value hedonic price indexes from this data that suggests that housing prices fell nearly 40 percent in nominal dollars between 1929 and 1932 with little recovery by 1940. These results contrast with Shiller which found that nominal housing values in 1940 nearly recovered to the levels found in 1929.

Working Paper

Heritage overlays are commonly employed across cities as a way to maintain the “character” of a neighbourhood. Heritage overlays in Melbourne vary in scope, but typically require additional permitting requirements in order to alter the existing structure. Qualitative household surveys are mixed on whether a heritage overlay benefits households. On the one hand, heritage may preserve the character of a neighbourhood, yet on the other, it limits redevelopment. Recent theoretical models generally suggest that while property prices fall, the overall welfare falls. Empirically, we find that heritage overlays are associated with a 7-13% increase in property values, yet the results vary by zone, the restrictions present within the heritage overlay as well whether the property is in the middle or outskirts of a heritage overlay. We find the presence of heritage overlays are nearly offset if the overlay maintains internal alteration controls. Moreover, properties that are surrounded by heritage overlays, but are not under a heritage overlay themselves benefit from the protection. Lastly, we find that the benefit of heritage overlays is strongest in lower density neighbourhoods and that residential properties in industrial and commercial zoning do not benefit from heritage protections.

Working Paper