DIFF command results interpretation

[Cyril Bergovic]

I ran a difference-in-differences model using the command DIFF (Villa, 2016) in Stata/IC 15.1 for Windows on a sample of 726 firm-years. The command I used is the following:

diff Y, t(treated) period(post) cov(controls) cluster(ID)

where Y is the dependent variable, post is a dummy equal to one after the event, controls is a set of control variables and standard errors are clustered at firm ID level. I obtain the following output:

DIFFERENCE-IN-DIFFERENCES WITH COVARIATES
DIFFERENCE-IN-DIFFERENCES ESTIMATION RESULTS
Number of observations in the DIFF-IN-DIFF: 726
	Before	After
Control	230	210	440
Treated	147	139	286
	377	349
Outcome var.	Y	S. Err.	|t|	P>|t|
Before
Control	0.051
Treated	0.046
Diff (T-C)	-0.005	0.003	-1.47	0.147
After
Control	0.053
Treated	0.065
Diff (T-C)	0.012	0.009	1.39	0.168
diff-in-diff	0.017	0.007	2.28	0.025**
R-square: 0.15
* Means and Standard Errors are estimated by linear regression
**Clustered Std. Errors
**Inference: *** p<0.01; ** p<0.05; * p<0.1

My difference after and my difference before are not significant but my diff-in-diff term is.

How can I interpret this result? If the difference is not significant after does it mean I cannot conclude on any link between treatment and outcome variable?

I thank you for your help.

References
Villa, J.M., 2016. diff: Simplifying the estimation of difference-in-differences treatment effects. Stata Journal 16, pp. 52-71

[Cyril Bergovic]

Please let me know if you need any further information.

[Cyril Bergovic]

Can anyone help me with this question ?