Supply Chain Network and Macroeconomic Expectations of Firms

Our intervention has two treatment arms and one control arm.
1—Treatment 1 (Mean Treatment): The information provided in this arm concerns the first moment of the GDP forecast, i.e., the first moment of future economic growth.
2—Treatment 2 (Uncertainty Treatment): The information provided in this arm concerns the second moment of the GDP forecast, i.e., the second moment of future economic growth.
3- Control: No information is provided in this arm.

2024-07-08

2024-09-30

1- Expectations. Macroeconomic expectations come from the baseline surveys (priors) and endline surveys (posteriors).

2-Economic Decisions. The second outcome examines firm's economic decisions (relating to the prices, employment, investment and wages).

The baseline survey collects the priors using the following questions:
1- What do you think will be the annual growth rate of real GDP in New Zealand in twelve months? Please provide an answer in percentage terms.
Answer: …………… \% per year

Could you provide us with an approximate range of what you think annualized real GDP growth in New Zealand will be over the next 12 months?
Answer: Real GDP growth over the next 12 months will be between ....... % per year (lowest forecast) and ...... % per year (highest forecast)

This question is directed exclusively to firms that received the treatment.
Please let me know what you perceive as the most pessimistic, the most likely, and most optimistic real GDP growth rate for New Zealand over the next 12 months. What do you think the lowest annualized real GDP growth rate might be for this time period, what do you think the most likely might be, and what do you think the highest might be? (please provide an answer as % per year).
Lowest real GDP growth rate: ……….. \% per year
Most likely GDP growth rate: ……….. \% per year
Highest real GDP growth rate: ……….. \% per year

The endline survey collects this information using the following question:
Please let me know what you perceive as the most pessimistic, the most likely, and most optimistic real GDP growth rate for New Zealand over the next 12 months. What do you think the lowest annualized real GDP growth rate might be for this time period, what do you think the most likely might be, and what do you think the highest might be? (please provide an answer as % per year).
Lowest real GDP growth rate: ……….. \% per year
Most likely GDP growth rate: ……….. \% per year
Highest real GDP growth rate: ……….. \% per year



The baseline survey collects the predictions of how the firm plans to change the prices over the next three months (prior to the information provided about the GDP forecast).

Over the next 3 months, by how much (in \% changes relative to current level) do you expect to change:
a) The price of your main product: ……………. \%
b) Investment in capital goods: ……………. \%
c) Employment at your firm: ……………. \%
d) Average wages: ……………. \%

The endline surveys ask the respondent the actual actions taken by the firm using the following question:

Over the last 3 months, by how much (in \% changes) did you change:
a. The price of your main product: ……………. \%\\
b. Investment in capital goods: ……………. \% \\
c. Employment at your firm: ……………. \% \\
d. Average wages: ……………. \%

1- Supply chain Dependent Expectation. We include the reasons behind the manager's responses on endline expectations to be related or unrelated to supply chain considerations.

2- Communication. The endline survey includes questions to quantify the frequency of communication for different topics between firms.

Using our firm-firm pairs, we randomly assigned the sample into three groups. The first group receives information about the first moment of GDP forecast (average GDP in Year 2025). We call this Treatment 1. In contrast, the second group receives information about the second moment of GDP forecast (uncertainty around GDP forecast for Year 2025). We call this Treatment 2. The last group receives no information and this is referred as the control group. Within each treated group, we further randomized whether the firm receives direct (main firm) or indirect information (linked firm). For the later group, we call those firms as untargetted treated firms because of their proximity to firms that receive direct information. By doing this, we are able to create 4 sub-groups for the treated sample: (1) Treatment 1 given to customer firms; (2) Treatment 1 given to supplier firms; (3) Treatment 2 given to customer firms; (4) Treatment 2 given to supplier firms.

We conduct the study in two waves in NZ with firms. The first is a baseline survey, during which the treated firms receive the information at the end. We then return to these firms after three months, allowing sufficient time for the information to disseminate.

Not available

Randomization done in office by computer

Firm-Firm Pair

No

Around 1000 Firm-Firm Pairs in Wave 1 (baseline)
Around 500 Firm-Firm Pairs in Wave 2 (endline) [due to expected attrition]

Around 2000 firms (suppliers + customers)

Around 150 firm-firm pairs in Treatment 1 (where the customer is the main firm)
Around 150 firm-firm pairs in Treatment 1 (where the supplier is the main firm)
Around 150 firm-firm pairs in Treatment 2 (where the customer is the main firm)
Around 150 firm-firm pairs in Treatment 2 (where the supplier is the main firm)
300 - 400 firm-firm pairs in control.

With the sample size of 150 (N1=75 treated and N2=75 control pairs), significance (alpha) equal to 5% and power (1-kappa) equal to 80%, the minimum detectable effect (MDE) is 0.46 SD. When we vary the sample size to 200, MDE=0.398 SD.