Technical, allocative and economic efficiencies

Summary

In this tutorial, the speaker explains how to estimate technical, allocative, and overall efficiencies using R. The process includes importing datasets, manipulating data with the dplyr package, and defining models for efficiency estimation. The speaker walks through steps like selecting relevant variables, calculating technical efficiency, and using input prices for cost efficiency. Additionally, the tutorial covers how to compute allocative efficiency by dividing cost efficiency by technical efficiency. The speaker also demonstrates how to group and summarize efficiencies by city, providing insights into factors influencing these efficiencies. The session concludes with a brief mention of total factor productivity estimation.

Takeaways

• 😀 Set the working directory and import the dataset using appropriate functions like `read.csv()` or `fdat`.
• 😀 Use the `select()` function to manipulate and filter necessary columns for efficiency estimation, focusing on production inputs like labor, materials, and capital.
• 😀 Export the manipulated dataset to Excel using the `write_xls()` function for further analysis or sharing.
• 😀 Estimating technical efficiency (TE) requires defining a model with orientation (input or output) and returns to scale (constant returns to scale).
• 😀 The technical efficiency results are stored in the dataset and can be accessed using `f$te` in the console for further review.
• 😀 For cost efficiency (OE), input prices are defined using the `t()` function to transpose the price data for inputs like labor, materials, and capital.
• 😀 Cost efficiency is calculated using the `model_profit()` function, which accounts for the prices of inputs and outputs in the model.
• 😀 Allocative efficiency (AE) is derived by dividing cost efficiency (OE) by technical efficiency (TE) and can be stored in the dataset.
• 😀 Grouping the efficiency results by variables like city allows for better comparison, helping to identify which city performs best in terms of technical or cost efficiency.
• 😀 After grouping, export the summarized efficiency data (e.g., mean values) by city to Excel for further comparison and analysis.
• 😀 The second stage of analysis can involve deeper insights, such as panel regression, to investigate inefficiencies and potential drivers of performance.

Q & A

What is the first step in the efficiency estimation process?

The first step is to set the working directory and import the dataset using the `import()` function, which helps load the required data for further analysis.

Which libraries are required for data manipulation in the script?

The necessary libraries for data manipulation are `dplyr` for handling and cleaning the data, and `writexl` for exporting results to Excel.

How do you clean the data to select relevant variables for efficiency estimation?

You clean the data by selecting only the relevant variables, such as firm, labor, materials, capital, and sales, while eliminating non-numeric variables like city or industry type that are not needed for efficiency calculations.

How is technical efficiency (TE) calculated in this script?

Technical efficiency (TE) is calculated by defining a model using the `model_basic()` function with a custom return to scale. Then, the `efficiency()` function is used to calculate the TE, which is added to the dataset.

What is the purpose of defining input prices in the process of estimating cost efficiency?

Defining input prices is crucial for estimating cost efficiency, as it accounts for the price of the inputs used in production. This is done by transposing the columns for input prices and including them in the overall efficiency model.

What function is used to calculate overall efficiency (OE), and how is it different from technical efficiency?

The `model_profit()` function is used to calculate overall efficiency (OE). Unlike technical efficiency, which focuses on the input-output relationship, OE incorporates the prices of inputs, reflecting the efficiency of cost management in the production process.

How is allocative efficiency (AE) calculated?

Allocative efficiency (AE) is calculated by dividing the overall efficiency (OE) by the technical efficiency (TE). This allows you to understand how efficiently resources are allocated in the production process.

How can the efficiencies be grouped by city or other variables for analysis?

Efficiencies can be grouped by city or any other variable of interest using the `group_by()` function from the `dplyr` package. After grouping, the `summarize()` function is used to calculate the mean values of technical, allocative, and overall efficiencies for each group.

What should you do if you want to export the grouped results to Excel?

You can export the grouped results to Excel using the `write_xls()` function, after grouping the efficiencies by city or other variables. This allows for further analysis and comparison in an external tool like Excel.

Why is it important to sort the results by efficiency in descending order?

Sorting the results in descending order helps identify the best-performing entities, such as the cities with the highest technical and overall efficiency. This enables more meaningful comparisons and insights into what drives efficiency.