That's a huge assumption. I could easily propose a half dozen other models each with a different conclusion. You would need to look at the data.