Make interest rate on government debt a function of the debt to GDP ratio

[jdebacker]

Since the risk premium of governments may vary with the amount of debt they accumulate, this PR updates the fiscal.get_r_gov function to allow it to vary with the debt to GDP ratio in a quadratic manner. The new specification adds to parameters to the function determining r_gov. The new function is:
$r_{gov,t}= (1-\tau_{d,t})r_{t} - \mu_d + \beta_1 \frac{D_t}{Y_t} + \beta_2 \left(\frac{D_t}{Y_t}\right)^2$

In the code $\beta_1$ maps to r_gov_DY and $\beta_2$ maps to r_gov_DY2 to avoid confusion with the rate of time preference.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 60.86957% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 73.01%. Comparing base (8cf609a) to head (bc9dc7a).

Files with missing lines	Patch %	Lines
ogcore/TPI.py	0.00%	9 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1037      +/-   ##
==========================================
- Coverage   73.19%   73.01%   -0.19%     
==========================================
  Files          21       21              
  Lines        5137     5143       +6     
==========================================
- Hits         3760     3755       -5     
- Misses       1377     1388      +11     

Flag	Coverage Δ
unittests	73.01% <60.86%> (-0.19%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
ogcore/SS.py	78.80% <100.00%> (-2.21%)	⬇️
ogcore/__init__.py	100.00% <100.00%> (ø)
ogcore/fiscal.py	97.97% <100.00%> (+0.10%)	⬆️
ogcore/TPI.py	33.55% <0.00%> (-0.08%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[jdebacker]

@rickecon, This PR is ready for your review.

[rickecon]

@jdebacker. Thanks for submitting this PR. Great update. Here is one suggestion and two questions.

• This is a relatively small addition to OG-Core. I recommend that this be a patch version update to 0.14.5, rather than a minor version update to 0.15.0 (see Semantic Versioning).
• In your fiscal.py, the get_r_gov() function takes DY_ratio as an argument. Why in your TPI.py in line 707 does r_gov take 0 as the DY_ratio value?
• Have you run an example locally with r_gov_DY>0 and r_gov_DY2>0?

[rickecon]

@jdebacker. Why is the DY_ratio argument in this line equal to 0? You have the D and the Y series above.

[jdebacker]

I've updated this to take the initial guess at the debt to GDP path (although since these are initial guesses, they shouldn't affect the model solution, but make affect the time to convergence).

[jdebacker]

Still a couple tests that I expect to pass failing locally:

FAILED tests/test_TPI.py::test_run_TPI_full_run[Baseline] - assert False
FAILED tests/test_TPI.py::test_run_TPI_full_run[Baseline, delta_tau = 0] - assert False

In both cases, it's a diff in bmat_splus1 (much larger in the $\delta_\tau$ = 0 case)

[jdebacker]

All tests pass locally:

tests/test_SS.py .........................................               [  6%]
tests/test_TPI.py ......................                                 [ 10%]
tests/test_aggregates.py .........................................       [ 17%]
tests/test_basic.py ....                                                 [100%]
tests/test_demographics.py ................                              [100%]
tests/test_elliptical_u_est.py .......                                   [100%]
tests/test_execute.py .                                                  [100%]
tests/test_firm.py ..................................................... [ 76%]
................                                                         [100%]
tests/test_fiscal.py .........................                           [100%]
tests/test_household.py ................................................ [ 94%]
...                                                                      [100%]
tests/test_output_plots.py ............................................. [ 95%]
..                                                                       [100%]
tests/test_output_tables.py ..............                               [100%]
tests/test_parameter_plots.py ........................................   [100%]
tests/test_parameter_tables.py .......                                   [100%]
tests/test_parameters.py ..............                                  [100%]
tests/test_pensions.py .................................                 [100%]
tests/test_run_example.py ..                                             [100%]
tests/test_run_ogcore.py .                                               [100%]
tests/test_tax.py ...................................                    [100%]
tests/test_txfunc.py ..............................                      [100%]
tests/test_user_inputs.py .........                                      [100%]
tests/test_utils.py .................................................... [ 62%]
...............................                                          [100%]

@rickecon This PR is ready for review.

[rickecon]

@jdebacker. I don't understand the 4 errors in test_fiscal.py. They are all "ValueError: assignment destination is read-only". It looks like we are writing to a NumPy array that is read only. But I don't understand why it would be read only.

[jdebacker]

@rickecon I see the issue, Pandas 3.0.0 (released yesterday) now treats Numpy arrays created from df.values as readonly. Will update to make these arrays from copies.

[rickecon]

@jdebacker. This looks great. Thanks for adding those .copy() commands in test_fiscal.py. Merging.