Performance cost of using Partials in Razor views

I like using partials in my Razor views. It makes the code nice and clean. But is there any significant performance cost of using partials? I've created a simple test. It shows that using partials is much slower.

test.cshtml:

@{
   var stopwatch = new System.Diagnostics.Stopwatch(); 
   stopwatch.Start();         
   for(var i=0; i<1000; i++) {
      var str = "my string #" + i;        
      @Html.Partial("_MyPartial",str)  // replace with @str on second test
      <br />        
   }
   stopwatch.Stop();
   <br />
   @: Time elapsed (ms): @(stopwatch.ElapsedMilliseconds)
}

_MyPartial.cshtml:

@Model

The code with the partial executed in 340ms, while the inline @str code was showing 0 or 1 ms. It is really shocking for me because it means I should get rid of all my cute partials, at least of those in loops.

If anyone wants to confirm or criticize my experiment, you are very welcome.