The Northridge earthquake: hospital capacity and pricing

BRIEFING NOTE
The Northridge earthquake as a natural experiment in market structure


Source:
Ennis, S. (1995) "The Northridge Earthquake: A Natural Experiment in Market Structure", University of California, Berkeley, Department of Economics Working Paper No. 95-244, November. https://digicoll.lib.berkeley.edu/record/206045/files/b22239617_C049557631.pdf

Executive Summary:
This working paper examines the impact of the 1994 Northridge earthquake on hospital prices in the Los Angeles area, using this natural disaster as a natural experiment. The author studies how the sudden reduction in hospital capacity, due to structural damage at several facilities, affected price competition. The study focuses particularly on Santa Monica Hospital, where a significant decrease in beds was followed by a notable increase in prices, suggesting a link between capacity and hospital pricing power. This analysis highlights the value of exogenous shocks in understanding the dynamics of markets.
Keywords:
Northridge Earthquake
Impact on hospital prices
Capacity reduction
Natural experiment
Local hospital market
Capacity and price
Northridge Earthquake: Impact on Hospital Prices

Main Theme:
This paper examines the impact of an exogenous shock to hospital capacity—in this case, structural damage caused by the 1994 Northridge earthquake—on prices for hospital services in the Los Angeles area. The study uses this natural disaster as a "natural experiment" to identify the effect of reduced capacity on price competition, overcoming the classic problem of endogeneity of market variables. Prior theoretical research is ambiguous about its prediction of the effect of capacity on pricing, depending on whether modelled as a one-shot or repeated game.

The Northridge Earthquake as an Exogenous Shock:

The earthquake of January 17, 1994, caused significant structural damage to 18 hospitals in the Los Angeles area.

Santa Monica hospitals, particularly St. John's and Santa Monica Hospital, suffered particularly severe damage, resulting in a significant reduction in capacity.

St. John's Hospital saw its capacity reduced from 501 to 262 beds after its reopening in October 1994, while Santa Monica Hospital saw its capacity reduced from 367 to 144 beds immediately after the earthquake.

The author argues that this earthquake constitutes an exogenous shock to hospital capacity, because it is a natural event independent of the decisions of market players.

Identifying the Effect of Capacity on Prices:

The exogenous reduction in capacity at Santa Monica allows us to study its impact on prices by avoiding the simultaneity problem where prices and capacity are usually determined jointly.

The study uses hospitals located in areas not affected by the earthquake as a control group to establish expected prices in the absence of capacity changes in Santa Monica.

The main methodology is to compare the price evolution at Santa Monica Hospital before and after the earthquake with that of similar hospitals that did not experience capacity reductions.

Local Hospital Market and Control Groups:

The author emphasizes that hospital markets are generally local, which allows for the construction of relevant control groups composed of unaffected hospitals operating under similar market and regulatory conditions.

Analysis of patient flows by ZIP code reveals that Santa Monica hospitals compete more intensely with each other than with other hospitals further afield, suggesting a somewhat distinct local market.

Hospitals similar to Santa Monica Hospital and St. John's Hospital are identified based on characteristics such as patient volume, occupancy rate, patient demographics, and complexity of cases treated.

Empirical Results for Santa Monica Hospital:

The regressions show a positive and significant coefficient for the post-earthquake dummy variable in the price models for Santa Monica Hospital, suggesting an increase in prices after the capacity reduction.

“ The dummy variable from this test has a positive and significant coefficient, suggesting that the capacity reduction due to the earthquake may have had a positive effect on prices. “

Even when using different regional price indices or comparing with similar hospital groups, the positive and significant effect of the earthquake on prices at Santa Monica Hospital persists.

Actual patient-day revenue values at Santa Monica Hospital after the earthquake are consistently higher than forecast values, initially more than 25% higher and always at least 13% higher than the forecast prices .

Empirical Results for St. John's Hospital:

Unlike Santa Monica Hospital, the estimated models do not show a significant effect of the earthquake on St. John's Hospital's patient-day revenue.

The regressions for St. John's Hospital have low explanatory power, and the coefficient on the post-earthquake dummy variable is not significantly different from zero.

The author notes a high variance in revenue per patient-day at St. John's Hospital, which could complicate the interpretation of the results.

Interpretation and Limits:

The significant increase in prices at Santa Monica Hospital after the earthquake could be attributed to the reduction in capacity, which would have diminished the credibility of insurers' threats to move patients to other hospitals.

The lack of a significant effect on prices at St. John's Hospital introduces necessary caution in interpreting the results for Santa Monica Hospital.

Limitations of the study include the small number of post-earthquake observations, the possibility of omitted variables (although regional employment does not appear to have been fundamentally altered), and the need for future research to confirm these findings in other settings.

“There are several reasons we might doubt findings from this data. First, the number of post-earthquake observations is small.”

Comparison with Previous Studies:
◦
The results of this study are consistent with the findings of a previous cross-sectional analysis (Melnick et al., 1992) which suggested that hospitals with high occupancy rates charge higher prices. However, exogeneity is clearer, allowing better identification of effects that, on a theoretical basis, are unclear.

Methodological Contribution:

The article highlights the usefulness of "natural experiments", such as natural disasters, to generate exogenous shocks to market variables and facilitate the identification of causal relationships, particularly in the context of industrial structure.

“ This paper both analyzes a problem and illustrates a method. The problem is to estimate the extent to which lowering capacity might increase prices. The method is to take advantage of natural disasters, such as earthquakes, as generators of exogenous shocks to market variables, such as capacity.”

Conclusion:

The study suggests that the exogenous reduction in hospital capacity in Santa Monica following the Northridge earthquake led to a significant increase in prices for hospital services there. Although the results for St. John's Hospital are less conclusive, the study highlights the potential of natural shocks to analyze the impact of market structure on price competition and underscores the need for future research to confirm these initial findings. The methodology employed offers a valuable approach to overcoming the identification challenges encountered in market analysis.


Paper Summary Initial Draft By NotebookLM